{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bu notebooktan tam verim almak için <a href=\"https://nbviewer.org/github/VolkiTheDreamer/PythonRocks/blob/master/Machine%20Learning/Supervised/E2E_Classification_ML_Main_Concepts_Part_I.ipynb\">Uçtan Uca ML notebookuyla</a> birlikte incelenmesini öneriyorum. Ve ayrıca <a href=\"https://github.com/VolkiTheDreamer/PythonRocks/tree/master/mypyextj\">kendi modüllerimden oluşan paketimi</a> de indirmeniz gerekmektedir. Bunun sebebi, sizi gereksiz kod kalabalığı ile yormak istemeyip dikkatinizi sadece buradaki ana konuya toplamaktır. Kodları isterseniz ayrıca inceleyebilirsiniz. İndirmek için şu adresi kullanın: https://minhaskamal.github.io/DownGit/#/home.\n",
    "\n",
    "Ayrıca sık sık <a href=\"Linear%20Regression.ipynb\">Lineer Regresyon</a> notebookuma da referansta bulunacağım."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gerekli kütüphanelerin import edilmesi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "all warnings will be shown\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from mypyext import dataanalysis as da\n",
    "import sweetviz as sv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#preprocessors\n",
    "from mypyext import ml\n",
    "from sklearn.model_selection import train_test_split,cross_val_score,cross_val_predict,StratifiedKFold,RepeatedKFold,RepeatedStratifiedKFold\n",
    "from sklearn.experimental import enable_iterative_imputer\n",
    "from sklearn.impute import SimpleImputer,IterativeImputer,KNNImputer\n",
    "from sklearn.preprocessing import StandardScaler,MinMaxScaler,RobustScaler\n",
    "from sklearn.preprocessing import OneHotEncoder, OrdinalEncoder, LabelEncoder\n",
    "from category_encoders import OrdinalEncoder as COE\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.base import TransformerMixin,BaseEstimator\n",
    "from sklearn.pipeline import Pipeline,make_pipeline\n",
    "from sklearn.preprocessing import FunctionTransformer\n",
    "from sklearn.model_selection import GridSearchCV,RandomizedSearchCV\n",
    "from sklearn.feature_selection import VarianceThreshold,SelectKBest, chi2, f_classif, mutual_info_classif,RFE,RFECV"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Şablon"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Algoritmaları anlatırken aşağıdaki şablona göre gitmeye çalışacağım:\n",
    "\n",
    "- **Teori**: \n",
    "    - Algoritmanın kısa tanımı ve nasıl çalıştığı\n",
    "    - Çözülmeye çalışılan problemin detayı\n",
    "    - Manuel implementasyon: Hazır kütüphane olmadan kendimiz nasıl kodlardık(Bu her zaman bulunmayabilir)    \n",
    "    - Varsayımlar/Ön kontroller: Bu algoritmayı kullanmak için hangi şartların sağlanması gerekir\n",
    "    - Önemli hususlar: bunun içinde algoritmanın nelere duyarlı olduğu, özellikle nelere dikkat edilmesi gerektiği ve (hyper)parametrelerde özellikle dikkat edilmesi gereken hususlar olacak. Ayrıca alogritmanın özellikle iyi olduğu alanlar varsa bunlara da bu başlıkta değineceğiz\n",
    "    \n",
    "- **Kod pratiği/Örnekler**: Konuyu pekişitrmek adına bir veya birkaç örnek. Konular ilerledikçe örnek sayısı azalabilir. Sırayala okunacağını düşünerek konununu pekişmesi adına ilk başlarda daha fazla örnek olabilir.\n",
    "- **Kaynaklar**: En sonda toplu olarak bulunmakla birlikte, yer yer aralarda da gerektiği durumlarda linkler verilecektir."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Teori"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Nedir? Nasıl çalışıyor? Nerde/Niçin kullanılır?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logistic Regresion(LogReg), bir sampleın hangi sınıfa/labela ait olduğunu tahmin etmeye çalışan, en temel algoritmalardan bir tanesidir. Bunu, lineer regresyondaki(LinReg)'nin aksine lineer bir fonksiyonuyla değil de, non-lineer bir fonksiyon olan **logit** fonksiyonu kullanarak hesaplar."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "LogReg tanıtılırken çoğunlukla şöyle anlatılır: *LogReg, adında Regression geçmesine rağmen bir regresyon algoritması değildir, classification algoritmasıdır*. Sonra başkaları bunu düzeltir ve der ki *Hayır, aslında bu bir lineer modeldir, zaten sklearn içinde de linear_model modülü içinde yer alır*. Evet, doğrudur, bu bir lineer modeldir, tıpkı LinReg'da olduğu gibi betalar yani katsayılar vardır ve lineer bir desicion boundarysi vardır. Devam edelim. Sonra başka bir kaynak der ki, *Evet her ne kadar arka planda bu bir lineer modelse de, sonuçta bir classification algoritmasıdır.* \n",
    "\n",
    "Hadi bu karmaşaya bi son verelim. Bu algoritmanın yaptığı, bir veri setindeki kayıtları sınıflandırmak değil, bunların belirli bir sınıfa ait olma olasılığını bulmaktır; sınıflandırma kısmı biz son kullanıcılara kalır. Belirlediğimiz thresholda göre sınıfı biz belirlemiş oluruz. Özetle LogReg, `classification(sınıflandırma) amaçlı kullanılan linear bir modeldir ve özünde bir regresyon algoritmasıdır.` <a href=\"https://towardsdatascience.com/how-are-logistic-regression-ordinary-least-squares-regression-related-1deab32d79f5\">Şuraya</a> da bi bakın derim."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logistic kelimesi `logit` fonksiyonundan geliyor, bildiğimiz lojistikle(taşımacılık, mal temini v.s) bi alakası yok. Bu model de <a href=\"Linear Regression.ipynb\">LinReg</a> gibi inputların ağırlıklarını hesaplar, ancak bunları doğrudan çıktı olarak vermek yerine bunları bu logit fonksiyonuna göndererek 0-1 arasında bir olasılık hesaplatır. Olasılıkları doğrudan kullanmak yerine log'unun kullanılması bir olayın olma olasılığı ile olmama olasılığı arasında simetri yakalama amacıyladır. Aşağıdaki videolardan birinde bunu daha iyi anlayacaksınız. Ben de biraz aşağıda bu konulara değineceğim."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "LogReg, en temel algoritmalardan biri olup, aynı zamanda çoğu durumda ilk başta denenmesi gerekenlerdendir. Birçok kullanım alanı olmakla birlikte tipik olarak \"mail spam mi değil mi\", \"bu hücre kanser hücresi mi değil mi\", \"bu işlem fraud işlemi mi değil mi\" gibi binary classification(yani 1/0, True/False, Yes/No) problemlerinde kullanılır, bununla birlikte multi-class classification da yapılabilmektedir."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Evet şimdi tanımlarda biraz daha derine inmeden, aşağıdaki videoları izleyelim, ama mutlaka izleyelim."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-02-06T17:04:43.902564Z",
     "start_time": "2022-02-06T17:04:43.418594Z"
    },
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from IPython.display import YouTubeVideo\n",
    "YouTubeVideo('yIYKR4sgzI8')\n",
    "# YouTubeVideo('ARfXDSkQf1Y')\n",
    "# YouTubeVideo('8nm0G-1uJzA')\n",
    "# YouTubeVideo('vN5cNN2-HWE')\n",
    "# YouTubeVideo('BfKanl1aSG0')\n",
    "# YouTubeVideo('xxFYro8QuXA')  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"mark\">\n",
    "DİKKAT:</div><i class=\"fa fa-lightbulb-o \"></i>Dedik ki, bu algoritma sadece bir olasılık hesaplar. Sklearn default olarak bu olasılık %50 üzerindeyse bir instanceı ilgili sınıfa sokar. Ama bu çok doğru bi yaklaşım değildir. Başka sık yapılan bir hata da threshold olarak prior probability kullanmak, yani bir sınıfın toplam küme içindeki oranını kullanmaktır, ki bu da çok doğru bi yaklaşım değildir. Burda thresholdu iş problemine göre kendimiz belirlemeliyiz. Bunla ilgili detay açıklamaları ve örneği <a href=\"E2E_Classification_ML_Main_Concepts_Part_I.ipynb\">Uçtan Uca ML projesi notebookunda</a> bulabilirsiniz."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Çözülecek problem"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Burdaki soru şudur: Elimizdeki prediktörleri kullanarak bir instance'ın hangi sınıfta olma olasılığını modelleyebilir miyiz?\n",
    "\n",
    "Logistic Regression'da bu soruya vereceğimiz cevap bizi Deep Learning'i(DL) de anlamaya yaklaştırackatır. Göreceksiniz, LogReg'in mantığı DL'in mantığına ne kadar benziyor. Zira burada da elimizdeki feature değerlerini bir aktivasyon fonksiyonuna(sigmoid) sokma var, DL'de de; keza burda da optimizasyon yöntemi olarak Gradient Descent kullanılıyor, DL'de de. DL ile ilgilenmeyi düşünüyorsanız LogReg'i ve onun temelinde yer alan logit/sigmoid fonksiyonlarını, cost functionları ve gradient descent'i anlamak çok önemli."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logit ve sigmoid fonksiyonları"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Şimdi, x'leri kullanarak bir Y olayı olur mu bunun olaslığını bulmak istiyoruz. İlk olarak klasik LinReg benzeri bir eşitlikle ilerlemeyi deneyelim."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$P(Y=1|X)= p=\\beta_0 + \\beta_1x$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Burda bi sıkıntı var, zira eşitliğin sağ tarafı herhangi bi değer(-inf ile +inf arasında) alabilir, halbuki bize pozitif bi değer lazım, zira olasılık denen şey pozitif bir değerdir. O zaman yukarıdaki gibi değil de exponasniyel bi denklem mi arask., $e^{\\beta_0+\\beta_1x}$ gibi. Bu işimizi görür gibi ama bunda da bi sıkıntı var ki, bu da artı sonsuza kadar gidiyor, bize 0-1 arası değer lazım. O zaman bunu aşağıdaki gibi ifade edelim. Böylece hem pozitif hem 0-1 arası değer elde ederiz.\n",
    "\n",
    "$$\\large p=\\frac{e^{\\beta_0+\\beta_1.x}}{1+e^{\\beta_0+\\beta_1.x}}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Burada karşımızda `odds` kavramı çıkıyor. O da şöyle hesaplanır:<br><br>\n",
    "\n",
    "$$\\large odds=\\frac{\\text{bi olayın olma olasılığı}}{\\text{bi olayın olmama olasılığı}}=\\frac{P(Y=1)}{P(Y=0)}=\\frac{p}{1-p}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Yukardaki kutudaki değerleri yerine koyarsak, odds şöyle çıkar:\n",
    "\n",
    "$$\\large e^{\\beta_0+\\beta_1.x}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Şimdi örnek probabilityler(p) üzerinden gidelim."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-02-04T18:05:19.989924Z",
     "start_time": "2022-02-04T18:05:19.260558Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\volka\\AppData\\Local\\Temp/ipykernel_18528/1686613978.py:3: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  odds=p/notp\n"
     ]
    }
   ],
   "source": [
    "p=np.array([0, 0.0001, 0.001, 0.01, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.95, 0.999, 0.9999, 1]) #eventin olma olasılığı\n",
    "notp=np.array(p[::-1]) #eventin olmama olasılığı\n",
    "odds=p/notp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-02-04T18:05:36.356393Z",
     "start_time": "2022-02-04T18:05:34.889517Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(p,odds,\"o-\")\n",
    "plt.xlabel(\"p\")\n",
    "plt.ylabel(\"odds\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Burda olayın olma olasılığı arttıkça odds ratio'nun da astronomik seviyelere çıktığını görüyoruz. Yani bi asimetriklik var. Bunu düzeltmek lazım. Nasıl? İki tarafın da log'unu alarak. odds'un logunu aldığımızda karşımıza çıkan fonksiyon `logit fonksiyonu` oluyor işte. Bu sayede karşımızda çıkan katsayıların yorumlanması daha kolay olacaktır, zira artık elimizde lineer bir denklem vardır.\n",
    "\n",
    "$$\\large odds=\\frac{p}{1-p} \\;\\;\\;\\;\\;\\text{-----log alalım---->}\\;\\;\\;\\;\\; ln(odds)=ln(\\frac{p}{1-p}) = ln(e^{\\beta_0+\\beta_1x})=\\beta_0+\\beta_1x$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Farkettiyseniz logit fonksyionun sonucu yine p'nin kendisi oluyor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-02-04T18:07:59.914810Z",
     "start_time": "2022-02-04T18:07:58.720341Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\volka\\AppData\\Local\\Temp/ipykernel_18528/2653175882.py:1: RuntimeWarning: divide by zero encountered in log\n",
      "  logit=np.log(odds)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "logit=np.log(odds)\n",
    "plt.plot(p,logit,\"o-\")\n",
    "plt.xlabel(\"p\")\n",
    "plt.ylabel(\"logit\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Artık elimizde lineer bir denklem var. X, 1 birim arttıkça Y'nin logiti $\\beta_1$ kadar değişir.\n",
    "\n",
    "Şimdi son olarak sınıf olasılıklarını bulmak için bu logit fonksiyonun da aşağıdaki gibi tersi alınır, ki buna da `sigmoid function` denir. Burda x ve y arasındaki ilişki non-lineer olmuştur."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-02-04T18:10:58.320815Z",
     "start_time": "2022-02-04T18:10:57.129366Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sigmoid=1/(1+np.exp(-logit))\n",
    "plt.plot(logit,sigmoid,\"o-\")\n",
    "plt.xlabel(\"logit\")\n",
    "plt.ylabel(\"sigmoid\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Şimdi $\\beta_0+\\beta_1x$ denklemimizi düşündüğümüzde burdaki betaları nasıl bulacağız? Esas çözülecek problem bu. En küçük kareler yöntemini(OLS) kullanamayız, çünkü ilişki lineer değil. Bu sefer `Maximum Likelihood Estimation(MLE)` metodunu kullanacağız ve bunu maksimize eden katsayıları bulmaya çalışacağız. Yani karşımızda yine bir optimizasyon görevi var. \n",
    "\n",
    "Şimdi hesapları yapalım:\n",
    "\n",
    "$$\\large P(Y=1\\;|\\;x)= \\frac{1}{1+e^{-(\\beta_0+\\beta_1x)}} = F_\\beta(x)$$\n",
    "<br>\n",
    "\n",
    "$$\\large P(Y=0\\;|\\;x)= 1 - F_\\beta(x)$$\n",
    "\n",
    "<p>Tek bir instance(gözlem) için likelihood tahmini genelleştirilmiş haliyle şöyle olur</p><br>\n",
    "\n",
    "$$\\large P(Y=y\\;|\\;x)= [F_\\beta(x)]^y.[1-F_\\beta(x)]^{1-y}$$\n",
    "\n",
    "Burda y'ler gerçek değerler iken $F_\\beta$'lar tahmini değerlerdir. Tüm instanceler için bu olasılıklar aşağıdaki gibi çarpılır ve nihai Likelihood fonksiyonu elde edilir.<br>\n",
    "\n",
    "$$\\large  L = \\prod_{i=1}^N[F_\\beta(x)]^{y_i}.[1-F_\\beta(x)]^{1-{y_i}}$$\n",
    "\n",
    "Bu çarpım işlemi işleri karmaşıklaştırır, üstelik bilgisayar belleği için de yüktür. Biz yine her iki tarafın logunu alalım, o yüzden fonksiyonumuzun adı artık Likelihood değil `Log Likelihood` olur. Log aldığımızda neler oalcağını bi hatırlayalım,\n",
    "\n",
    "$$log(a^b)=b.log(a)\\;\\; ve \\;\\;log(a.b)= log(a)+log(b)$$\n",
    "\n",
    "ve sonuç\n",
    "\n",
    "$$\\large  log(L) = LL = \\sum_{i=1}^N[{y_i}.log(F_\\beta(x)^i) + (1-{y_i).log(1-F_\\beta(x)^i)]}$$\n",
    "\n",
    "veya \n",
    "\n",
    "$$\\large  LL = \\sum_{i=1}^N[{y_i}.log(p_i) + (1-{y_i).log(1-p_i)]}$$\n",
    "\n",
    "\n",
    "Bu nihai fonksiyona **cross entropy** adı verildiğini de görebilirsiniz. Bunun binary case için özel hali ise **binary cross entropy** olup aşağıdaki gibidir.\n",
    "\n",
    "$$\\large  LL = \\sum_{i=1}^N{y_i}.log(p_i)$$\n",
    "\n",
    "Bi adımımız daha kaldı o da, maximize edilmeye çalışılan bu fonksiyonu minimize etmeye çalışmak. Bunun için de başına bi \"-\" konur. \n",
    "\n",
    "$$\\large  Negatif Log Likelihood = NLL = -\\sum_{i=1}^N{y_i}.log(p_i)$$\n",
    "\n",
    "Nihai fonksiyonumuz negatif işaretlidir, bu artık bir **cost function** olmuştur. **iterasyon** ile bu maliyet fonksiyonunu minimize etmeye çalışacağız. Bunun için de `Gradiend Descent(GD), Stokastik GD, Newton metodu` gibi metodlar var. Biz burada GD'ye bakacağız. \n",
    "\n",
    "Şimdi öncelikle GD öncesine kadarki olan kısmı, yani yukarıda anlattıklarımızı bir de şematik olarak görelim."
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)\n",
    "<p style=\"font-size:smaller;text-align:center\">Görsel: H.S Ölmez - Sabanci University</p>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Veya Kaggle master Kaan hocamızın <a href=\"https://www.kaggle.com/kanncaa1/deep-learning-tutorial-for-beginners#Logistic-Regression\">gösterdiği</a> gibi beta yerine weight anlamında w'ler de kullanılabilir, ki bu şematik gösterim Neural Networks(Sinir Ağları/Derin Öğrenme) anlatımında da karşımıza çıkacak."
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Gradient Descent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bunu LinReg notebookunda görmüştük, oraya tekrar bakabilirsiniz. Regresyon değil de classification bağlamında görmek için Kaan hocamızın yine yukarıdaki linkine bakabilirsiniz. Bunlara ek olarak aşağıdaki linklerden de gerek GD detayını gerek manuel implementasyonu görebilirsiniz.\n",
    "\n",
    "- https://towardsdatascience.com/logistic-regression-explained-and-implemented-in-python-880955306060\n",
    "- https://realpython.com/logistic-regression-python/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\"Bana kısaca sen anlat\" diyenler için şöyle özetleyeyim.\n",
    "\n",
    "- Öncelikle \"kaytsayıların ilk değerlerine ne verelim\" sorusuyla başlanır. Bunun için bazı teknikler var, diyelim ki 0.01 verdik ve $\\beta_0$(bias) için de 0 dedik.\n",
    "- x'ler ile betalar(weightler) çarpılır ve toplanır. Çıkan sonuç, bir aktivasyon fonksiyonu olan sigmoid fonksiyonuna sokulur. Diyelim ki eğitim setinde bir instance'ın classını 0(not-churn) tahminledik ve gerçekten de 0'mış(veya 1 dedik ve gerçekten 1 çıktı), o zaman kaybımız(`loss`) 0'dır. Bu işlemin adı **forward propagation**'dır.\n",
    "- Tüm instancelar için bu loss'ların toplamına da `cost` deniyor. Nihai amaç, cost'un minimize olması.\n",
    "- Sonra başa dönüp betalar ve bias güncellenir, ki buna da **backward propagation** denir. Güncelleme işlemi de türev alarak gradient descent yöntemiyle yapıyoruz, ta ki eğim(yani türev) 0 olana kadar."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cost function olarak LogLoss(binary cross entropi)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cost functionımız yukarıdaki negatif log-likelihood veya diğer adıyla binary cross entropi fonksiyonudur. Buna negative log loss da denmektedir. Yukarıda bahsedilen tüm proses boyunca bu metrik minimize edilmeye çalışılır.\n",
    "\n",
    "$$\\large  Negatif Log Likelihood = NLL = -\\sum_{i=1}^N{y_i}.log(p_i)$$\n",
    "\n",
    "Denklemden görüldüğü üzere, gerçek sınıf 1 iken eşitliğin ikinci kısmı uçuyor, 0 iken de ilk kısmı. Bu denklemin bir güzelliği de, yanlış tahmini ne kadar kendinden emin bi şekilde yaparsak(yani olasılığımız ile gerçeklik arasındaki fark ne kadar çok açıksa) bunun daha fazla cezalandırılmasıdır. Mesela aşağıdaki örneğe bakalım,"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "actual:1, True olasılığı:0.99, logloss:0.01, normalloss:0.01\n",
      "actual:0, True olasılığı:0.99, logloss:4.61, normalloss:0.99\n",
      "actual:1, True olasılığı:0.01, logloss:4.61, normalloss:0.99\n",
      "actual:0, True olasılığı:0.01, logloss:0.01, normalloss:0.01\n"
     ]
    }
   ],
   "source": [
    "prd=[0.99, 0.99, 0.01, 0.01] #tahmin olasılıkları\n",
    "act=[1, 0, 1, 0] #gerçek değerler\n",
    "\n",
    "for i in range(len(act)):    \n",
    "    logloss=-(act[i]*np.log(prd[i])+(1-act[i])*np.log(1-prd[i]))\n",
    "    normalloss=np.abs(act[i]-prd[i])\n",
    "    print(f\"actual:{act[i]}, True olasılığı:{prd[i]:.2f}, logloss:{logloss:.2f}, normalloss:{normalloss:.2f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Aşağıda daha detaylı bilgiler edinebilrsiniz ancak özet olarak şunu söyleyebiliriz. Regresyon analizlerinden genelde SSE(Sum of Squared Errors), classficationda ise Log Loss/CrossEntropy Loss optimize edilmeye çalışılır. Tabi classficationda classlara farklı ağırlıklara vererek bu cost functionları modifiye etmek de mümkündür. Bunun nasıl yapıldığını <a href=\"E2E_Classification_ML_Main_Concepts_Part_II.ipynb\">Uçtan uca ML projesinde</a> görebilirsiniz.\n",
    "\n",
    "Bu konu da önemli bir konu olup ilave okumalar yapmanızı öneririm\n",
    "\n",
    "- https://towardsdatascience.com/cross-entropy-negative-log-likelihood-and-all-that-jazz-47a95bd2e81\n",
    "- https://towardsdatascience.com/intuition-behind-log-loss-score-4e0c9979680a\n",
    "- https://towardsdatascience.com/linear-regression-using-gradient-descent-97a6c8700931\n",
    "- https://towardsdatascience.com/common-loss-functions-in-machine-learning-46af0ffc4d23\n",
    "- https://towardsdatascience.com/understanding-the-3-most-common-loss-functions-for-machine-learning-regression-23e0ef3e14d3\n",
    "- https://www.analyticsvidhya.com/blog/2019/08/detailed-guide-7-loss-functions-machine-learning-python-code/\n",
    "- https://www.analyticsvidhya.com/blog/2020/11/binary-cross-entropy-aka-log-loss-the-cost-function-used-in-logistic-regression/\n",
    "- https://www.data4v.com/log-loss-as-a-performance-metric/\n",
    "- https://medium.com/konvergen/cross-entropy-and-maximum-likelihood-estimation-58942b52517a\n",
    "- https://medium.com/@phuctrt/loss-functions-why-what-where-or-when-189815343d3f\n",
    "- https://algorithmia.com/blog/introduction-to-loss-functions\n",
    "- https://towardsdatascience.com/understanding-sigmoid-logistic-softmax-functions-and-cross-entropy-loss-log-loss-dbbbe0a17efb\n",
    "- https://towardsdatascience.com/understanding-binary-cross-entropy-log-loss-a-visual-explanation-a3ac6025181a\n",
    "- http://www.awebb.info/probability/2017/05/18/cross-entropy-and-log-likelihood.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Son olarak şunu da söylemekte fayda var. GridSearch içinde de scoring parametresine de accuracy/precision gibi metriclere ek olarak neg_log_loss da verebiliyoruz. Yani neg_log_loss hem Logistic Regresyonunu optimize etmeye çalıştığı bir **fonksiyondur** hem de bir **evaluation metriğidir**. Daha detay bilgi için <a href=\"E2E_Classification_ML_Main_Concepts_Part_I.ipynb\">Uçtan uca ML projesi(PartI)</a> içinde GridSearch bölümündeki Önmeli Husulara bakabilirsiniz."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Manuel Implementasyon"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Manuel implementasyonu Kaan hocamızın <a href=\"https://www.kaggle.com/kanncaa1/deep-learning-tutorial-for-beginners#Logistic-Regression\"> Kaggle sayfasında</a> bulabilirsiniz."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Varsayımlar"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- LinReg'in aksine Residual'ların normal dağılımı ve homoscdedasticity gerekmez\n",
    "- LinReg'in aksine prediktör ve target arasında Lineer ilişki gerekmez\n",
    "- LinReg'de olduğu gibi featurelar arasında multicollinearity, featureların yorumlanmasında sorun teşkil edebilir\n",
    "- LinReg'de olduğu gibi instanceların birbirinden bağımsız olması beklenir\n",
    "- LinReg'de olduğu gibi featureler arasın collinearity olmaması gerekir(Tahmin sonucunu değiştirmez, ama featureların önemini yorumlamada önemlidir)--> bunla ilgili kaynaklara <a href=\"Linear Regression.ipynb\">LinReg</a> notebookundan bakabilirsiniz.\n",
    "- Instance sayısı feature sayısının en az bi 10-15 katı olmalıdır"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Önemli husular"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Genel**\n",
    "- Çok kritik bi detay değil ama mülakatlarda çıkabilir diye tekrar belirtmekte fayda var: Sınıflandırma algoritması değildir, sınıflandırmada kullanılan lineer regresyon algoritma türü\n",
    "- fit çizgisi S şeklindedir, ama decision boundry lineerdir\n",
    "- Maximum Likelihood(MLE) maximize edilmeye çalışılır(veya negative log likelihood cost function minimize edilemesi)\n",
    "- Kategorik featureları  **onehot encode** etmeye gerek yoktur\n",
    "- Target'ı LabelEncode etmeye gerek yoktur\n",
    "- Data linearly separable durumdaysa, MLE,  sınıflar arasındaki ayrımı ortaya koymak içinde katsayıların gittikçe büyümesine neden olur. Bunu önlemek için **Penalty** kullanmak gerekir. lamda(sklearn:1/C) değeri deneme yanılmayla optimize edlir. **(penalty l1/l2 değil mi? sait hoca notlarına bakalım**\n",
    "- defaultu binary classfication içindir ama multi_class paramteresi multinomial yapılarak mutlcilass tahminleme yapılabilir.\n",
    "- SGDClassfication: SGD, genel olarak bir optimizasyon yöntemidir. Bu anlamda, SGDClassifer da, regularizasyon içeren bir linear modelin SGD(Stochastic Gradient Descent) ile optimize edilmiş halidir. sklearn'de LogReg için böyle bir classifer var: SGDClassifier(alpha=k, penalty='l2', loss='log') \n",
    "\n",
    "\n",
    "**Avantajlar**\n",
    "- computation comlexity:O(nd), hızlı eğitilir.\n",
    "- online/realtime kullanımı vardır.\n",
    "- Interpretability'si yüksektir(katsayılar aracılığıyla)\n",
    "\n",
    "**Dezavantajlar**\n",
    "- İyi bir optimizasyon elde etmek için yüksek  sayıda veriye ihtiyaç duyar, az veride başarılı olmayabilir\n",
    "- Outlierlara karşı duyarlıdır, dikkatlice ele alınması gerekir.\n",
    "- Scaling'e duyarlıdır(Tahmin sonucunu değiştirmez, yorumlamada önemli)\n",
    "- lineer decision boundry'si vardır. linearly sperable olmayan datalarda, ki çoğunlukla öyle oalcaktır, performansı düşüktür. O yüzden de diğer algorimtalra göre genelde daha düşük bir accuracy vardır. O yüzden çoğunlukla baseline model olarak seçilir."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Kod Pratiği"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data temini, analizi(EDA) ve preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Titanic verisetini inceleyecğeiz, bu birçok eğitimde anlatılan bir veri setidir, ve <a href=\"https://www.kaggle.com/competitions/titanic/overview\">Kaggle'da</a> da bulunmaktadır.\n",
    "\n",
    "Çeşitli bilgileri verilen yolcuların hayatta kalıp kalmadığı bilgisi var. Bu bilgileri kullnarak bir model oluşturacağız ve yeni gelen veri setindeki bir kişinin hayatta kalıp kalmadığnı tahmin etmeye çalışacağız.\n",
    "\n",
    "**Önemli not**: İki ayrı veri seti verilmiş durumda. train ve test diye. Böyle iki parça halinde verilen setlerde genelde test setinde label olmaz, ve bunu bizim tahmin etmemiz, sonra da sonuçları bir yere yüklememiz istenir. Gerçek değerleri biz bilmeyiz, onlar bu verisetini yaratanların elindedir. Bu örnekte de durum böyle. Bu durum biraz kafanızı karıştırabilir. Şöyle yapalım, burdaki test setini saha verisi olarak düşünün. Biz okuyacğımız train verisini ise elimizdeki ana veri gibi düşünüp, onu yine kendi içinde train ve teste ayıracağız."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-02-04T18:39:24.706251Z",
     "start_time": "2022-02-04T18:39:23.190631Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=3f1d5b26-5e9f-4487-9c32-b0359ad55994 style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('3f1d5b26-5e9f-4487-9c32-b0359ad55994').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Thayer)</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "3            4         1       1   \n",
       "4            5         0       3   \n",
       "\n",
       "                                                Name     Sex   Age  SibSp  \\\n",
       "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
       "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
       "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
       "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0      1   \n",
       "4                           Allen, Mr. William Henry    male  35.0      0   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  \n",
       "0      0         A/5 21171   7.2500   NaN        S  \n",
       "1      0          PC 17599  71.2833   C85        C  \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S  \n",
       "3      0            113803  53.1000  C123        S  \n",
       "4      0            373450   8.0500   NaN        S  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.read_csv(\"https://raw.githubusercontent.com/VolkiTheDreamer/dataset/master/Classification/titanic_train.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Survived kolonu bizim target kolonumuz,binary classification olacak. Kolonlarda şunlar açıklamaya ihtiyaç duyuyor, diğerleri zaten aşikar:\n",
    "\n",
    "- pclass: Ticket class (1 = 1st, 2 = 2nd, 3 = 3rd)\n",
    "- SibSp: # of siblings / spouses aboard the Titanic\n",
    "- Parch: # of parents / children aboard the Titanic\n",
    "- embarked: Port of Embarkation(C = Cherbourg, Q = Queenstown, S = Southampton)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-02-04T18:39:46.401053Z",
     "start_time": "2022-02-04T18:39:45.604458Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(891, 12)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## EDA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Genel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=1219f0bf-d81c-4395-982a-bf667d65aad6 style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('1219f0bf-d81c-4395-982a-bf667d65aad6').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Type</th>\n",
       "      <th>Nunique(Excl.Nulls)</th>\n",
       "      <th>#of Missing</th>\n",
       "      <th>MostFreqItem</th>\n",
       "      <th>MostFreqCount</th>\n",
       "      <th>First</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <td>int64</td>\n",
       "      <td>891</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Survived</th>\n",
       "      <td>int64</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>549</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pclass</th>\n",
       "      <td>int64</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>491</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Name</th>\n",
       "      <td>object</td>\n",
       "      <td>891</td>\n",
       "      <td>0</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>1</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sex</th>\n",
       "      <td>object</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>male</td>\n",
       "      <td>577</td>\n",
       "      <td>male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Age</th>\n",
       "      <td>float64</td>\n",
       "      <td>89</td>\n",
       "      <td>177</td>\n",
       "      <td>24.0</td>\n",
       "      <td>30</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SibSp</th>\n",
       "      <td>int64</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>608</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Parch</th>\n",
       "      <td>int64</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>678</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ticket</th>\n",
       "      <td>object</td>\n",
       "      <td>681</td>\n",
       "      <td>0</td>\n",
       "      <td>347082</td>\n",
       "      <td>7</td>\n",
       "      <td>A/5 21171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fare</th>\n",
       "      <td>float64</td>\n",
       "      <td>248</td>\n",
       "      <td>0</td>\n",
       "      <td>8.05</td>\n",
       "      <td>43</td>\n",
       "      <td>7.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cabin</th>\n",
       "      <td>object</td>\n",
       "      <td>148</td>\n",
       "      <td>687</td>\n",
       "      <td>B96 B98</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Embarked</th>\n",
       "      <td>object</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>S</td>\n",
       "      <td>644</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "                Type  Nunique(Excl.Nulls)  #of Missing  \\\n",
       "PassengerId    int64                  891            0   \n",
       "Survived       int64                    2            0   \n",
       "Pclass         int64                    3            0   \n",
       "Name          object                  891            0   \n",
       "Sex           object                    2            0   \n",
       "Age          float64                   89          177   \n",
       "SibSp          int64                    7            0   \n",
       "Parch          int64                    7            0   \n",
       "Ticket        object                  681            0   \n",
       "Fare         float64                  248            0   \n",
       "Cabin         object                  148          687   \n",
       "Embarked      object                    4            2   \n",
       "\n",
       "                        MostFreqItem  MostFreqCount                    First  \n",
       "PassengerId                        1              1                        1  \n",
       "Survived                           0            549                        0  \n",
       "Pclass                             3            491                        3  \n",
       "Name         Braund, Mr. Owen Harris              1  Braund, Mr. Owen Harris  \n",
       "Sex                             male            577                     male  \n",
       "Age                             24.0             30                     22.0  \n",
       "SibSp                              0            608                        1  \n",
       "Parch                              0            678                        0  \n",
       "Ticket                        347082              7                A/5 21171  \n",
       "Fare                            8.05             43                     7.25  \n",
       "Cabin                        B96 B98              4                      NaN  \n",
       "Embarked                           S            644                        S  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.super_info_() #extension metodum, kullanımı için: https://mvolkanyurtseven.medium.com/top-n-useful-python-tips-tricks-e3a163e56749"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ilk gözlemlerimiz:\n",
    "\n",
    "- null kolonlar var\n",
    "- düşük ve yüksek <a href=\"https://en.wikipedia.org/wiki/Cardinality\">kardinalitesi</a> olan kolonlar var\n",
    "- full cardinalitesi olan 2 kolon var(id ve name), bunları sileriz(Belki Name'den Mr/Mrs gibi ünvanları da alabiliriz ama şuan buna odaklanmayalım)\n",
    "- Ticket bilgisi anlamsız bi bilgi gibi duruyor, bunu da silebiliriz(Belki bundan bile bi anlam çıkartılabilir, ama yine bunu es geçelim)\n",
    "- ordinal, numerik ve kategorik türlerin hepsi var\n",
    "- Cabin bilgisinden anlamlı bir feature türetebilir miyiz(feature extraction) bi bakalım"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "#silinecekleri silelim\n",
    "df.drop([\"PassengerId\",\"Name\",\"Ticket\"], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Düşük kardinalitesi olanların değerlerine bakalım"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unique items in column Survived\n",
      "[0 1]\n",
      "\n",
      "Unique items in column Pclass\n",
      "[3 1 2]\n",
      "\n",
      "Unique items in column Sex\n",
      "['male' 'female']\n",
      "\n",
      "Unique items in column SibSp\n",
      "[1 0 3 4 2 5 8]\n",
      "\n",
      "Unique items in column Parch\n",
      "[0 1 2 5 3 4 6]\n",
      "\n",
      "Unique items in column Embarked\n",
      "['S' 'C' 'Q' nan]\n",
      "\n",
      "You may want to consider the numerics with low cardinality as categorical in the analysis\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['Survived', 'Pclass', 'Sex', 'SibSp', 'Parch', 'Embarked']"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "da.getColumnsInLowCardinality(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cabin bilgisine bakalım"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "B96 B98        4\n",
       "G6             4\n",
       "C23 C25 C27    4\n",
       "C22 C26        3\n",
       "F33            3\n",
       "F2             3\n",
       "E101           3\n",
       "D              3\n",
       "C78            2\n",
       "C93            2\n",
       "Name: Cabin, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Cabin.value_counts().head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "sanki ilk karakterlerini alıp gruplayabilriz gibi, belki de bunlar geminin belirli bir katını veya bloğunu gösteriyordur, ve belki atıyorum B ile başlayanların büyük çoğunluğu kurtulmuştur."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"CabinGrup\"]=df.Cabin.fillna(\"ZZZ\").apply(lambda x:x[0])\n",
    "del df[\"Cabin\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Şimdi veri türlerimizi belirleyelim."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(['Age', 'SibSp', 'Parch', 'Fare'],\n",
       " ['Pclass', 'Sex', 'Embarked', 'CabinGrup'],\n",
       " ['Pclass'],\n",
       " ['Sex', 'Embarked', 'CabinGrup'])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target=[\"Survived\"]\n",
    "nums=[\"Age\",\"SibSp\",\"Parch\",\"Fare\"]\n",
    "cats=list(df.columns).removeItems_(nums+target,False) #extension metodum\n",
    "ords=[\"Pclass\"]\n",
    "noms=cats.removeItems_(ords,False)\n",
    "\n",
    "nums,cats,ords,noms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visuals"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Korelasyonlar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 345.6x288 with 20 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df[nums],height=1, aspect=1.2);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Diagonala bakıldığında Age dışındakiler skewed görünüyor, bunlardan SiSp ve Parch zaten küçük sayılar bi log transformasyona gerek yok ama Fare'de gerekli gibi\n",
    "- Yine diagonaldakilerin kimisinde outlier da var gibi, boxplotla ayrıca bakarız\n",
    "- kolonlar arası collinearity yok görünüyor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#transformasyon öncesi\n",
    "df.Fare.hist(figsize=(5,2));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#transformasyon sonrası\n",
    "plt.figure(figsize=(5,2))\n",
    "plt.hist(np.log10(df.Fare+1));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tam normal dağılım olmadı ama yine de scaling için yeterli gibi."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Şimdi de korelasyonlara bakalım, bunun için dython kütüphanesinden faydalanacağız. Zira bu kütüphane ile hem numeric-numeric, hem numeric-kategorik, hem de kategorik-kategorik korelasyonlar tek bi fonksiyonla elde edilebilmektedir."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\volka\\appdata\\local\\programs\\python\\python38\\lib\\site-packages\\scikitplot\\plotters.py:33: DeprecationWarning: This module was deprecated in version 0.3.0 and its functions are spread throughout different modules. Please check the documentation and update your function calls as soon as possible. This module will be removed in 0.4.0\n",
      "  warnings.warn(\"This module was deprecated in version 0.3.0 and its functions \"\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from dython.nominal import associations\n",
    "corrdict=associations(df,nominal_columns=cats,numerical_columns=nums,figsize=(10,10))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Targetla en yüksek korelasyonu olan kolonlara bakalım"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sex          0.540200\n",
       "Pclass       0.336684\n",
       "CabinGrup    0.320034\n",
       "Fare         0.257307\n",
       "Embarked     0.173099\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_results=corrdict[\"corr\"] #dataframe\n",
    "da.getHighestPairsOfCorrelation(corr_results,\"Survived\",5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bu korelasyon değerlerinden feature selection aşamasında yararlanabiliriz."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x360 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[nums].plot(kind=\"box\", subplots = True,figsize=(6,5))\n",
    "plt.tight_layout();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6,4))\n",
    "for e,n in enumerate(nums):\n",
    "    plt.subplot(1,len(nums),e+1)\n",
    "    ch=sns.violinplot(y=n, data=df)\n",
    "plt.tight_layout()    \n",
    "plt.show();    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Age, Min:0.42, Max:80.0, Q1:    20.12, Q3:    38.00, IQR:    17.88, Q3+1,5*IQR:    64.81, Q1-1,5*IQR:    -6.69, Mean within the box:    69.77, Total Mean:    29.70, Outliers:11\n",
      "\n",
      "SibSp, Min:0, Max:8, Q1:     0.00, Q3:     1.00, IQR:     1.00, Q3+1,5*IQR:     2.50, Q1-1,5*IQR:    -1.50, Mean within the box:     4.37, Total Mean:     0.52, Outliers:46\n",
      "\n",
      "Parch, Min:0, Max:6, Q1:     0.00, Q3:     0.00, IQR:     0.00, Q3+1,5*IQR:     0.00, Q1-1,5*IQR:     0.00, Mean within the box:     1.60, Total Mean:     0.38, Outliers:213\n",
      "\n",
      "Fare, Min:0.0, Max:512.3292, Q1:     7.91, Q3:    31.00, IQR:    23.09, Q3+1,5*IQR:    65.63, Q1-1,5*IQR:   -26.72, Mean within the box:   128.29, Total Mean:    32.20, Outliers:116\n",
      "\n"
     ]
    }
   ],
   "source": [
    "da.outlierinfo(df,nums,imputestrategy=\"None\",thresh=0.25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "11 outliers exists in feature 'Age'\n",
      "46 outliers exists in feature 'SibSp'\n",
      "213 outliers exists in feature 'Parch'\n",
      "116 outliers exists in feature 'Fare'\n"
     ]
    }
   ],
   "source": [
    "da.outliers_IQR(df,nums,imputestrategy=\"None\",thresh=0.25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "46 outliers exists in feature 'SibSp'\n",
      "1 outliers exists in feature 'Parch'\n",
      "20 outliers exists in feature 'Fare'\n"
     ]
    }
   ],
   "source": [
    "da.outliers_IQR(df,nums,imputestrategy=\"None\",thresh=0.1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Evet hepsinde outlier var, özellikle Fare'de. SizSp ve Parch her ne kadar outlier gösterse de bunlar outlier olarak ele alınmamalı, çünkü bunlar kişi sayıs ve çok da bir uç değer yok aslında."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "12 outliers exists in feature 'SibSp'\n",
      "10 outliers exists in feature 'Parch'\n",
      "17 outliers exists in feature 'Fare'\n"
     ]
    }
   ],
   "source": [
    "da.outliers_zs(df,nums,imputestrategy=\"None\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2 outliers exists in feature 'Age'\n",
      "30 outliers exists in feature 'SibSp'\n",
      "15 outliers exists in feature 'Parch'\n",
      "20 outliers exists in feature 'Fare'\n"
     ]
    }
   ],
   "source": [
    "da.outliers_std(df,nums,imputestrategy=\"None\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sonuç olarak, thresholdu 0.1 olan IQR yöntemi ile ilerleriz."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Null kontrolü"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import missingno as msno\n",
    "msno.bar(df, figsize=(6,4));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\volka\\appdata\\local\\programs\\python\\python38\\lib\\site-packages\\missingno\\missingno.py:72: MatplotlibDeprecationWarning: The 'b' parameter of grid() has been renamed 'visible' since Matplotlib 3.5; support for the old name will be dropped two minor releases later.\n",
      "  ax0.grid(b=False)\n",
      "c:\\users\\volka\\appdata\\local\\programs\\python\\python38\\lib\\site-packages\\missingno\\missingno.py:141: MatplotlibDeprecationWarning: The 'b' parameter of grid() has been renamed 'visible' since Matplotlib 3.5; support for the old name will be dropped two minor releases later.\n",
      "  ax1.grid(b=False)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# da.nullPlot(df)\n",
    "msno.matrix(df,figsize=(8,5));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are no null-like values\n",
      "Don't forget to check for 0's manually\n"
     ]
    }
   ],
   "source": [
    "da.findNullLikeValues(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Target bazlı analizler"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Targetın dağılımına bakalım"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(x=df[\"Survived\"]);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.616162\n",
       "1    0.383838\n",
       "Name: Survived, dtype: float64"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Survived.value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Çok büyük olmamakla birlikte hafif bir imbalance sözkonusu."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Target bazında numeriklerin ortalama değerlerine bakalım."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "da.plotNumericsByTarget(df,\"Survived\",nums=nums,layout=(1,4),figsize=(12, 4))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fare ve Parch, target bazında oldukça farkediyor, bunların feature importance'ı önemli olacak gibi duruyor. Korelasyon analizinde Fare için bu durumu gözlemledmiştik zaten ancak Parch için aynı durum sözkonusu değil. (Outlier varsa sonuçları yorumlayı yanıltabilir, ki Fare'da oldukça büyük outlierlar vardı, o yüzden buraki sonuçlara şuan çok da güvenmemekte fayda var)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Kategorik kolonlar üzerinden baktığımızda ilgili kategorilerdeki her bir değer için Surviving olasılıklarına(prior probability) bakalım."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "da.plotTargetByCats(df, cats, \"Survived\", subplot_tpl=(2,2));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bu grafikleri şöyle okumak lazım. Öncelikle bunların percentage olduğunu unutmayın:\n",
    "\n",
    "- 1.sınıflarda hayatta klama olasılığı daha yüksek olmuş\n",
    "- Kadınlarda hayatta klama olasılığı daha yüksek olmuş\n",
    "- Cherborugdan binenler daha şanslıymış\n",
    "- T Cabin grubunda olanların hiçbiri hayatta kalamamış, sonra kabin numarası bilinmeyenler en şanssız iken, B,D ve E gruplarındakiler daha şanslıymış"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Şimdi de hayatta kalanların kategoriler bazındaki dağılımına bakalım."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14,8))\n",
    "da.plotPositiveTargetByCats(df, cats, \"Survived\", subplot_tpl=(2,4),pos_label=1);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Yorumlar\n",
    "\n",
    "Kurtulanların çoğunluğu;\n",
    "\n",
    "- Kadın\n",
    "- Southamptondan binenler \n",
    "- Kabin numarası bilinmeyenler"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Her bir kategorik feature ve numerik feature çifti için target'ın ortalama değerlerine bakalım."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plots for Age,\n",
      "----------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 466.56x172.8 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 466.56x172.8 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1399.68x172.8 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plots for SibSp,\n",
      "----------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 466.56x172.8 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 466.56x172.8 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1399.68x172.8 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plots for Parch,\n",
      "----------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 466.56x172.8 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 466.56x172.8 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABWUAAAClCAYAAAA0w585AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAdvElEQVR4nO3de5hkdX3n8fd3phmugotcRtSBVTQgXlBHd43rLSKixkt0TRDUHfVR0bhZVBwVjQJed9QoXlZjVrMqShIfycYEjQ/q4gU1Oqh4BcQIuM22zjgiDDDgMN/945xOV1dXX+tc6lS/X8/ze6bq1Ll869efrq751anficxEkiRJkiRJktSMNW0XIEmSJEmSJEmriYOykiRJkiRJktQgB2UlSZIkSZIkqUEOykqSJEmSJElSgxyUlSRJkiRJkqQGOSgrSZIkSZIkSQ1yUHaJImJTROwcdh2pl7lSHcyVqmamVAdzpTqYK9XBXKkuZktVM1PdsioGZSPi8Ig4NyJ+FhG3RsRkRHwuIp5Q8aH+Frj7Cuq7Q0ScHRE/jIibI2JHRFwaEWdGxCEV19iYiMiFWtv1DctctSci1kXEKyPiuz3P7ZsR8aKI2Lvt+oZhrtpRvjGZ7/Vqn7brG4aZak9E7BURryifz86IuCEivh8Rb4uIu7Vd3zDMVTv6Xqtuj4jrI2JrRLw5Ig5ru75hmat2le+vtpWvVwe1XU9VzFU7Bry3ujEivhURT2y7tqqYrXYs8L799LZrG5aZal5EXDxPnqbb1XUef6LOnY+CiDgKuAS4EXgNcBnFYPRjgA8CG6o6VmbeAtyyzPr+HfBV4I7AG4CtwPXA0cBzgOcBW+bZdl1m3rbyimt35wHLjgIuAj7abCnVMlftiYh1wOeBBwCvp3ie1wMPBl4OXAFc3FJ5QzFXrbsZuEf/wszc1UItlTBT7el5rToeOBv4GvArij7/Y+AVwOktlTcUc9W66deqAA6k+Pv3KuAFEfHIzPxJm8WtlLkaCU8Ffg78FjgF+ECr1VTAXLWu973VgcBLgAsi4ujM/EV7ZQ3PbLVu0Pv2G9oopCpmqjVPA9aVtw8GfgQ8Hfh6uez2Wo+emWPdgM8Ck8ABAx67Y8/tlwPfB24q1/+ffY9vAnYCTwKuBHYB/we4e/86PffPAn4InAz8jOKX638Dh/Ss84Fyv0fMU3/03L663OdHKML/KYpBzgQ29m2XwH8ub0+vcwrFfwp3AZcDJzb8s9gP+F7ZbxNtZ8NcdTNXwGZgT39t5WNrgAPbzoe56mSuZvXHuDQz1WqmXkXxJu4Biz23rjVzNXqvVcABZR9+ue18mKvu5aqnls8D/xV4NrC17UyYq27nqr8/ymUHlLU8vu1smK3xytY4NDM1En8HDymP/6imfu5jPX1BRBwMnAS8PzPnzJeRmdf33N1DccbKcRQBeAjw3r5N9qb4ROC5wEOBtRSf9MUCZRwF/AnwR8CJFGf3vbmsbw1F6M/LzOsGbZxlMnq8nCKUG4EzFzjuIFuA91CctXMR8A8RcZf5Vi5PQd+5SHv4Mo7/18BBwDMyc/cyax8Z5mqOpnN1KvCFzNw64HntycxOfkJqruZo+/Wq88zUHE1n6hTgosz87qAHBzy3TjBXc4zEa1X5s/gg8IiIOHS527fNXM3ReK4i4kjgUcDfABcAx0TE/ZdZ90gxV3O0+noVERMUfbeL4gzAzjJbc4zE38IuM1NzrJ5MNTX620ajCGcCf7SCbU8CbgXWlPc3lft6WM86R1KcBXNCzzr9nzbsAg7qWfZa4Kry9uHlPl/Wd+yvU3wCsRP4XM/yq4F/7Fv3KJb+acNrex5fQ/GpyZsW6IODKU5FX6jtu8T+fE35fO7fdi7MVbdzRfFVlXPbzoG5GrtcTffZzr729bazYaY6m6lb6HutAs7veW4/ajsj5qqTuZrVHwP6N4GHtJ0Tc9WtXPX0wT/13P8Y8L62s2Guupsr5r63up3iffzJbWfDbI1dtnYyz9/GrjQz1f7fwXI/jZ8pO8F4W+hTgNkrRvwBxcDhsRRnc66lmFdiPTD9ScAe4FvT22TmNRFxHXBv4Avz7PqazPxtz/3rgMUuxPAnzHyy0X8hkDlnBy7DN6ZvZOaeiPgXitoHyswdwI4hjgdAFJNSvxF4ZmZ2+lPRkrmarelcLbn/O8ZczdbG69XNFJ/G9rp1yH22yUzN1srfwD4vA/4ceD7wzIr33RRzNdso5Gra9M8ma9p/nczVbI3mqjwD6rkUU0RN+zjwdxFxRnZ3bnVzNVvb7632A04A/joibsjMzw657zaZrdnaztY4MFOzjdL7q1qN9fQFwE8p3pgeu9BK5dd1LgR+AjwDeBDFJMUwM+HvtOW+0f3dgO2n+30bxfwax8xaIfMXmXkVxST7/W7qu7+n/PfffokjYq9l1jhQRV+FuhfwSeBtmfmpKuoaAeZqCBXk6koW6fuOMldDqOgrK5mZV/W1Ll+EwkwNoaLXqv7nNlU+t19XUWNLzNUQav563b0p+uLqKmptmLkaQgW5OpHiAjKfiIjdEbEb+BzFxVyeXkWNLTFXQ6jhvdX3M/MvgC9TDCh1mdkaQl3v26uorUVmagg1v7+q1VgPypaj5Z8HXhoRB/Q/HhF3LG9upAjwyzLzG5l5JXDEgF2uoTitfHr7DeV6K7rKbWbuAf4WeFZE9H+qsFTbyn/v3LPs+HnW/Y/TNyIiKJ7LQrV/sNzXQm3eTz8i4kDgH4CvUJwZNBbM1RyN5opikP+EiNjY/0BErClz1znmao6mczV2zNQcTWfqfOCxg16rusxczTESr1Xlz+I0igt9bVts/VFjruZoOlfPp5hHtn+bvyof6yRzNcdIvF5RfIV6vxVsNzLM1hyjkq3OMlNzrJpMjfv0BQB/ClwCbI2IP6e4Sl0Aj6b4hG4DxacSa4DTI+ICigCcPmBfu4F3R8R/o5gr7l3Aj5j/9O+lOBN4JPDNiHgD8G2K+TiOo5hs/5cLbZyZt0TEN4FXRcTPKE5ff+s8q784Iq4EfgC8hGJekQ8ssO8VnwJe/uJ8guIP7iuAw2PunNI7MvO2lex/BJirGY3lqvRu4InAReVz+wrFJ3MPAM6geO4XD7H/NpmrGU3nCoqXrvUDlm/LzNuH3HdbzNSMpjP1LuAJwBci4mzgqxRnyB4NPIXiP6VdZa5mtP1adRDwYOBV5e0nD7nvNpmrGU2+Zz+UIjfPyMwf9j32YeAbEXGPzPzZSvY/AszVjLZfr/YFHgs8DjhnyP2OArM1o41sjSMzNWP1ZCpHYFLjuhvFSPx7gX+lmB/wOoqv5JzUs86fAZMUgf0i8McUp2sflT0TIVP8R+qn5X6+DBzds49NzJ0s+Yd9tcxap1x2IPAm4Mfl8W+h+AV8C3BYz3pXA2cMeH7HUvzy3kwR2oczeLLkUykmYt4FXAE8vsY+P7I85kLtUW1nw1x1K1c9te1N8R/Qy8rn9Rvgm8CLgHVtZ8NcdS9XzEyIP6gdXeexzdR4Zqo89jrglcB3y/puLY/9HuDubWfDXHUvV8x+rdpD8aHkd/qfV1ebuWrlPfvLgRuAvQc8FsA1wFvazoa56lauep5r7/up6eOeCaxtOxdmq/PZ6vSFvczUaGWqp77GL/QV5YE1xiLiKODnwIMzcyRP2Vb3mCvVwVypamZKdTBXqoO5Uh3MlepitlS11ZipsZ5TVpIkSZIkSZJGjYOykiRJkiRJktQgpy+QJEmSJEmSpAZ5pqwkSZIkSZIkNchBWUmSJEmSJElq0MgPyp500kkJ2GyD2oqZK9sCbUXMlG2RtiLmyrZIWxFzZVukrYi5si3SVsRc2RZoK2aubAu0FTFTtkXavBoblI2IvSPiwxFxTUTcGBHfi4jHL7bd9u3bmyhPq4y5UtXMlOpgrlQHc6U6mCvVwVypDuZKVTNTWqkmz5SdAH4BPBI4CHgd8HcRcVSDNUiSJEmSJElSqyaaOlBm3gSc1bPonyLi58CDgKubqkOSJGmlNm/ezNTUFOvXr2fLli1tlyNJkiSpo1qbUzYiDgfuBfxowGMvjIitEbF127ZtzRensWSuVDUzpTqYq9E2NTXF5OQkU1NTbZeyLOZKdTBXqoO5Uh3MlapmplSFVgZlI2Iv4BPARzPz8v7HM/NDmbkxMzceeuihzReosWSuVDUzpTqYK9XBXKkO5kp1MFeqg7lS1cyUqtD4oGxErAE+DtwGvLTp40uSJEmSJElSmxqbUxYgIgL4MHA48ITM/F2Tx5ckSZIkSZKktjU6KAt8ADgWOCEzb2n42JIkSZIkSZLUusamL4iII4EXAccDUxGxs2ynNlWDJEmSJEmSJLWtsTNlM/MaIJo6niRJkiRJkiSNosYv9CVJkiRJkiRJq5mDspIkSZIkSZLUIAdlJUmSJEmSJKlBDspKkiRJkiRJUoMclJUkSZIkSZKkBjkoK0mSJEmSJEkNclBWkiRJkiRJkhrkoKwkSZIkSZIkNchBWUmSJEmSJElqkIOykiRJkiRJktQgB2UlSZIkSZIkqUEOykqSJEmSJElSgxyUlSRJkiRJkqQGOSgrSZIkSZIkSQ1yUFaSJEmSJEmSGuSgrCRJkiRJkiQ1aKLtAiRJkoZ17Tn3beQ4u3ccDEywe8c1jRxzw+t/UPsxJEmSJDXPM2UlSZIkSZIkqUEOykqSJEmSJElSgxyUlSRJkiRJkqQGOSgrSZIkSZIkSQ1yUFaSJEmSJEmSGjSx3A0iYj/geOAw+gZ1M/OCasqSJEmSJEmSpPG0rEHZiDgBOB+404CHE1hbRVGSJEmSJEmSNK6WO33BucCFwF0zc01fc0BWkiRJkiRJkhax3OkLjgKenJnX1VCLJEmSJEmSJI295Z4pewnwe3UUIkmSJEmSJEmrwaJnykbEA3vufhB4R0QcAfwA+F3vupn5nWrLkyRJkiRJkqTxspTpC7ZSXMQrepZ9aMB6XuhLkiRJkiRJkhaxlEHZf197FZIkSZIkSavA5s2bmZqaYv369WzZsqXtciS1ZNFB2cy8polCJEmSJEmSxt3U1BSTk5NtlyGpZcu60FdEvDkiThuw/LSIeGN1ZUmSJEmSJEnSeFrWoCzwbOC7A5ZfCjxn+HIkSZIkSZIkabwtd1D2MGDbgOW/Bg4fvhxJkiRJkiRJGm/LHZS9Fnj4gOWPAP7v8OVIkiRJkiRJ0nhb9EJfff4SeFdErAO+VC57DPBW4L9XWZgkSZIkSZIkjaNlDcpm5jsj4hDgPcC6cvFtwLmZuWWx7SPipcAm4L7A+Zm5aVnVSmNk8+bNTE1NsX79erZsWfTXR5IkSZIkSWNiyYOyETEBnAi8A3gTcO/yoZ9k5s4l7ua6ctvHAfsuo05p7ExNTTE5Odl2GZIkSZIkSWrYkgdlM3N3RFwAHJOZvwa+vdyDZeYFABGxEbjrcreXJElS9/ltEUnqHl+7VQdzpdVsuXPKXgYcDVxdfSkzIuKFwAsBNmzYUOehtIqYK1XNTKkOS83Vg175saZKatSlb39O2yWMpVF7vfLbIuNh1HKl8WCuRleXX7vN1ejqaq7MlKqwZpnrnwW8MyKeGhF3i4iDe1tVRWXmhzJzY2ZuPPTQQ6varVY5c6WqmSnVwVyNtkP22cPh++7mkH32tF3Kspgr1cFcqQ7mSnUwV6qamVIVlnum7IXlvxcA2bM8yvtrqyhKkiRpFJ1xv+vbLkGSJEnSGFjuoOyja6lCkiRJkiRJklaJZQ3KZuaXhzlYREyUx1wLrI2IfYDdmbl7mP1KkiRJkiRJUlcs90xZACLiCGADsK53eWZ+ZZFNXwe8oef+s4CzKeaqlSRJkiRJkqSxt6xB2XIw9pPAIyjmkJ2eS3bagnPKZuZZOAArSZIkSbXavHkzU1NTrF+/ni1btrRdjiRJ6rNmmeu/G7gduDdwM/Bw4BnAT4CTKq1MkiRJkrQiU1NTTE5OMjU11XYpkiRpgOVOX/BI4ImZeXlEJLAtMy+JiFuBNwIXVV6hJEmSJEmSJI2R5Z4puy+wvby9AzisvP1j4H5VFSVJkiRJkiRJ42q5g7KXA8eUt78HnBYRRwJ/CkxWWJckSZIkSZIkjaXlTl9wLrC+vH0O8M/AM4Fbgf9SYV2SJEmSJEmSNJaWNCgbEfsBbweeCuwVEScCfwYcRXHm7LWZuX3eHUiSJEmSJEmSgKVPX3A2sAm4EDgfeCzwgcy8OTO/44CsJEmSJEmSJC3NUqcveBrw/Mz8G4CI+ARwSUSszczba6tOkiRJkiTN69pz7tvIcXbvOBiYYPeOaxo75obX/6CR40hSG5Z6puzdgK9O38nMbwG7gSPqKEqSJEmSJEmSxtVSz5RdC9zWt2z3MraXJKlRmzdvZmpqivXr17Nly5a2y5EkSZIk6d8sdVA1gPMi4taeZfsAfxURN08vyMwnV1mc1LQHvfJjjR3rDttvZC1w7fYbGznupW9/Tu3HWCkHz1SHqakpJicn2y5DkiRJ6pxxnRbDKTE0SpY6KPvRAcvOq7IQdYODZ6qDg2eSJEmSJGk1WdKgbGY+t+5C1A0OnkmSJEmSJEnDWeqFviRJkiRJkiRJFXBQVpIkSZIkSZIatNQ5ZSVJkiRJWlW8poYkqS4OykqSJAmAB73yY40c5w7bb2QtcO32Gxs55qVvf07tx5A0nrymhiSpLg7Kjolrz7lvI8fZveNgYILdO65p7JgbXv+DRo4jSZIkSZIkNcFBWUmSJKlFfj1akiRp9XFQVpIkSWqRX4+WJElafRyUXQLPXpAkSZIkrWaH7LMH2F3+O36amlcdnFtdUsFB2SXw7AVJXeAHSJIkSarLGfe7vu0SJGmsOCgraV7jegG5cb14XFc+QDJXkiRJkqTVzkFZLcu4f2VFkiRJkiRJqlunB2WbmvPF+V5m+JUVSZIkSZIkaThr2i5AkiRJkiRJklaTTp8pK0mSJEld4tzqkiQJHJSVJEmSJHWMU9lJkrrOQdkl2LNu/1n/SlUwV5IkjTbPaJQkqV5eTFyrmYOyS3DTPU9suwSNIXMlSZIkSVrNvJi4VjMv9CVJkiRJkiRJDXJQVpIkSZIkSZIa5PQFkqSx5PxUkiRJkqRR5aCspNaN++CZVwduh/NTSZIkSZJGlYOyklrn4JkkrS571u0/619JkiRptWl0UDYiDgY+DJwIbAdek5mfbLIGSZIkteume57YdgmSJElSq5o+U/b9wG3A4cDxwIURcVlm/qjhOiRJkqSRMO7T+EiSJGmuxgZlI2J/4OnAfTJzJ/C1iPgM8Gzg1U3VIUmSJI0Sp/GRJElafdY0eKx7Absz88qeZZcBxzVYgyRJkiRJkiS1KjKzmQNFPBz4VGau71n2AuDUzHxU37ovBF5Y3v094IpGilzYIRTz4Gp0+mJ7Zp601JXN1cgblb5Ycq7MVCeMSn+Yq/EyKv1hrsbLqPSHuRovo9If5mp8jEpf+H/B8TIqfeFr1XgZlf6YN1dNDso+ALgkM/frWfYK4FGZ+aRGihhCRGzNzI1t1zEK7Ivq2Jcz7Itq2I+z2R/VsB9nsz+qYT/OZn9Uw36czf6ohv04w76ojn05w76ohv04Wxf6o8npC64EJiLinj3L7g94kS9JkiRJkiRJq0Zjg7KZeRNwAXBOROwfEQ8DngJ8vKkaJEmSJEmSJKltTZ4pC/ASYF/gV8D5wIszsytnyn6o7QJGiH1RHftyhn1RDftxNvujGvbjbPZHNezH2eyPatiPs9kf1bAfZ9gX1bEvZ9gX1bAfZxv5/mhsTllJkiRJkiRJUvNnykqSJEmSJEnSquagrCRJkiRJkiQ1yEHZRUTEwRHx9xFxU0RcExGntF1TWyLipRGxNSJujYj/1XY9XWauZpirapip2cxVNczVDDNVHXM1w1xVx1zNMFfVMVczzFV1zNUMc1UNMzVbl3I10XYBHfB+4DbgcOB44MKIuKxDFyir0nXAm4DHUVywTStnrmaYq2qYqdnMVTXM1QwzVR1zNcNcVcdczTBX1TFXM8xVdczVDHNVDTM1W2dy5YW+FhAR+wO/Ae6TmVeWyz4OTGbmq1strkUR8Sbgrpm5qe1aushcDWauVs5Mzc9crZy5GsxMDcdcDWauhmOuBjNXwzFXg5mr4ZirwczVypmp+XUhV05fsLB7Abung126DDiupXo0HsyVqmamVAdzpTqYK9XBXKkO5kp1MFeqmpnqMAdlF3YAcEPfst8Cd2ihFo0Pc6WqmSnVwVypDuZKdTBXqoO5Uh3MlapmpjrMQdmF7QQO7Ft2IHBjC7VofJgrVc1MqQ7mSnUwV6qDuVIdzJXqYK5UNTPVYQ7KLuxKYCIi7tmz7P7Aap0sWdUwV6qamVIdzJXqYK5UB3OlOpgr1cFcqWpmqsMclF1AZt4EXACcExH7R8TDgKcAH2+3snZExERE7AOsBdZGxD4RMdF2XV1jrmYzV8MzU3OZq+GZq9nMVDXM1WzmqhrmajZzVQ1zNZu5qoa5ms1cDc9MzdWlXDkou7iXAPsCvwLOB16cmav1E4fXAbcArwaeVd5+XasVdZe5mmGuqmGmZjNX1TBXM8xUdczVDHNVHXM1w1xVx1zNMFfVMVczzFU1zNRsnclVZGbbNUiSJEmSJEnSquGZspIkSZIkSZLUIAdlJUmSJEmSJKlBDspKkiRJkiRJUoMclJUkSZIkSZKkBjkoK0mSJEmSJEkNclBWkiRJkiRJkhrkoGxHRcTFEfG+Bo6zMyI21X0ctc9MqQ7mSnUwV6qDuVIdzJXqYK5UNTOlOpirxTkoO4SIODQi/kdEXB0Rt0bELyPiixHx2AYO/zTgNQ0cRw0yU6qDuVIdzJXqYK5UB3OlOpgrVc1MqQ7marRNtF1Ax30a2A94PnAVcBjwSOBOK91hRKzLzNsWWy8zd6z0GBppZkp1MFeqg7lSHcyV6mCuVAdzpaqZKdXBXI2yzLStoAF3BBI4YYF1rgbO6Ft2MfC+vnXOAj4CXA98Cvg68M6+7Q4EbgGe1r8f4C3ApQOO/3XgPT33nwv8GNgFXAm8DFjT8/jR5X53AVcAfwjsBDa13d+roZkpm7kyV11p5spmrsxVV5q5spkrc9WFZqZs5mp15srpC1ZuZ9meHBH7DLmvlwOXAxuBM4HzgJMjovfn83SK0F04YPvzgAdGxDHTCyLi7sBDy8eIiBdQ/BK8HjgWeAXwKuAl5eNrgL+nmNLiocDzKH7p9h7yuWnpzJTqYK5UB3OlOpgr1cFcqQ7mSlUzU6qDuRp1bY8Kd7lRBG4HRei+AbwD+A89j1/N0j5x+Me+de4E3AY8pmfZF4APLbCf7wBv7Ln/OuCKnvvXAs/uO87pwI/L2ycCtwMbeh7/TxSfqmxqu69XSzNTNnNlrrrSzJXNXJmrrjRzZTNX5qoLzUzZzNXqy5Vnyg4hMz8NHAE8Cfgc8PvANyPizGXuamvffn8N/DNwKkBEHAE8mvLTg3mcB5zSc/9U4BPl9ocCdwP+Moqr0u2MiJ3A24B7lOsfC0xm5rU9+/gXYM8yn4uGYKZUB3OlOpgr1cFcqQ7mSnUwV6qamVIdzNVoc1B2SJm5KzMvysxzMvP3gQ8DZ0XEOopgRN8mew3YzU0Dlp0HPL08xfxk4BfAVxco5XzgyIh4aEQ8EDiGmV+G6Z/zacDxPe0+wHELPkE1zkypDuZKdTBXqoO5Uh3MlepgrlQ1M6U6mKvR5aBs9X4MTAD7ANuAO08/UAb1mHm26/eZ8t8/pPj04JNZnps9SGb+P+BL5bqnAt/IzH8tH/slcB1wj8y8qr+Vu/gJcJeIuFvPbh+CGRkFZkp1MFeqg7lSHcyV6mCuVAdzpaqZKdXBXI2IibYL6KqIuBPFFec+AnwfuJFiwuPNwBcz84aI+BLwvIj4DEXQX8sS+zwzd0XEpynm2Lg/8OwlbHYe8E6KeT3e3PfYG4D3RsT1wGcpPvl4IHCXzHwrxdwflwMfi4iXAfsC7wJ2L6VeDc9MqQ7mSnUwV6qDuVIdzJXqYK5UNTOlOpirDsgRmNi2i43i6m5vAb4N/Aa4Gfgp8BfAweU6B1Kcnv1bYJLiinEXM3fC5DPmOcYfUExY/J0Bj83aT7nsAIpTym8D7jRgm2dSTKy8q6z5a8DJPY/fC/gycGv5XJ5McaW+TW3392poZspmrsxVV5q5spkrc9WVZq5s5spcdaGZKZu5Wp25ivJJSZIkSZIkSZIa0Nl5FyRJkiRJkiSpixyUlSRJkiRJkqQGOSgrSZIkSZIkSQ1yUFaSJEmSJEmSGuSgrCRJkiRJkiQ1yEFZSZIkSZIkSWqQg7KSJEmSJEmS1CAHZSVJkiRJkiSpQQ7KSpIkSZIkSVKD/j9SBA67EQHWXgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1399.68x172.8 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plots for Fare,\n",
      "----------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 466.56x172.8 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 466.56x172.8 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1399.68x172.8 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "da.plotTargetForNumCatsPairs(df,nums,cats,\"Survived\",2.4,0.9)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Son olarak da targetın diğer kategoriler bazında ortalama numerik değerlerine bakalım"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plots for Age,\n",
      "----------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plots for SibSp,\n",
      "----------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plots for Parch,\n",
      "----------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Plots for Fare,\n",
      "----------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAS8AAAClCAYAAAANgnvSAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAV00lEQVR4nO3de5gcVZnH8e8vgSRcAoEEua2BBeWOAQxowBVExcgjGLmTiCIrKJhduQkKiJGLulzUBbyhi8gGXEQBQQQFMREhgEEEhXA3XALBxCGYe4S8+8epSXqanp7umenuqunf53nq6emqU1XvzOl5q+tU1TmKCMzMimZQqwMwM+sNJy8zKyQnLzMrJCcvMyskJy8zKyQnLzMrJCevHJE0TdJlTdjPIklHN3o/Vh/Xf32cvEpI2kjStyXNlrRc0suSfiPp/U0K4SDgC03aV0NJ2lnSdElLJc2RdLYktTqualz//UPSMElXSnpY0j8lTWvEftZoxEYL7GfA2sC/A08BbwL2Bkb2ZaOShkTEip7KRURHX/aTF5LWA24HfgfsDmwH/BBYDFzcwtB64vrvH4OBZcBlwP7AiIbsJSI8pacMRgABvK+HcrOBU8vmTQMuKyszBbgCWABcB9wDXFy23nrAUuCg8u0AXwEeqLD/e4BLSt5/AniU9GF5AjgJGFSy/C3ZdpcBjwMfAhYBRzfwb3k88A9grZJ5ZwFzALW6rl3/ja3/sngvA6Y1Yts+bVxtUTYdKGlYP2zvZOAxYCxwBjAVOEJS6d/8YNKH6pYK608FdpO0XecMSVsB47JlSDqW9CE/G9geOAU4HTghWz4IuIHUPDAOOIb0TzW0WuCSJmXtItWmSVU2MQ64KyKWlsz7FbAZsGW1fbeQ63/1fvpa/83R6iNenibSh6mD9IGaAVwEvKOszGxqO/LeXFZmJLACeG/JvDuAy6ts54/AuSXvzwIeL3n/HHBU2X5OBB7Nft4PeB0YXbL8XaRvGN0eeYHhpCN2tWl4lfV/DVxRNm90tt9xra5n139j679sW/7m1QwR8TPSt4MDgFuBPYF7JZ3Ri83NLNv234HbgEkAkjYD3kN2FO3GVGBiyftJwNXZ+hsBbwa+V3pEBL4GbJ2V3x6YExHPlWzjPmBltcAjYmFEPNXDtLD6r188rv9VsRai/p28ykTEsoi4PSLOiYg9gf8BpkgakhVZCZRfNVuzwqYWV5g3FTg4Oy05AngeuKtKOD8GtpA0TtJupIbvzg97Z919GtilZNoJ2LHKNnvUD6cNc4GNy+ZtXLIst1z/xTlt9NXGnj1K+jsNI33tnwds2rkw+yBuBzxYw7ZuAr5PajSdBFwT2XfrSiLiJUl3ZmWXAzMi4pls2cuSXgS2joirutnELGBzSW+OiOezeXvQ80HrJtIRupqXqyybAfyXpGERsSyb937gRdIpVZG4/iurVv9N4eSVkTSSdFXoCuBhYCGpsfU04DcR8Y+s6J3AMZJuIn2Qz6TGv2NELJP0M1LbxRjgqBpWm0q6vWAFcH7Zsi8Bl0paAPyS9A1gN2DziPgqqU3lMeAqSScBawHfAF7rIc6FpN+/t67JYrtS0nnANsDngS9X+2dtJdd/lzj7Wv9I2gEYAowC1pW0S7btP/Vlu100oiGtiBPpCsxXgD8ArwBLgCeBrwMblpRbj/R1/lXSpf8TqNxge2o3+9mX1GD6xwrLumwnm7cu6RRkBTCywjpHkhp2l2Vx/x44omT5NsB00pH7SeBAmnCpHNiZdJ/XMuAl0j9aLm+TcP035O85O/s9u0z9uQ9lOzIzKxQ32JtZITl5mVkhOXmZWSE5eZlZITl5mVkhDejkNX78+DdcqvVUyKlXXP8DZqpoQCev+fPntzoEayHX/8A2oJOXmQ1cfjzIzLp12mmnMXfuXDbZZBMuuOCCVofThZOXmXVr7ty5zJkzp9VhVOTTRjMrJCcvMyskJy8zKyQnLzMrJCcvMyskJy8zKyQnLzMrJCcvMyskJy8zK6SmJi9JkyXNlLRc0pUl87eUFGXjwn2xZPlQSVdI+oekuZJObmbcZpY/zX486EXgPOADpGGYyo2IiErDMk0B3gpsAWwC/FbSoxFxW6MCNbN8a+o3r4i4PiJuBP5e56ofB86NiFciYhZp4M6j+zk8MyuQvLV5PSvpBUk/lDQKQNIGpBGKHyop9xDdDGku6bjs1HTmvHnzGh+x5Yrrv33kJXnNB3YnnRa+HRgOXJ0tWzd7fbWk/KtZmTeIiMsjYmxEjN1oo40aFK7lleu/feSiS5yIWATMzN6+LGky8JKk4aTRfSGNVLys5Oc+DUduZsWWl29e5Tr7rR4UEa+QhosfU7J8DPBI06Mys9xo9q0Sa0gaBgwGBksals17h6RtJQ2SNBK4BJgWEZ2nilcBZ0naQNJ2wLHAlc2M3czypdnfvM4ClgKfBz6a/XwWsBVwG+lU8C/AcuDIkvW+BDwNPAtMBy70bRJm7a2pbV4RMYV0z1YlP66y3nLgmGwyM8ttm5eZWVW5uNqYd3keQcWsXTl51SDPI6iY9adZ59/Z5f2KjqWrXkuXbX/mvk2NqxKfNppZITl5mVkhOXmZWSG5zcusB75gk09OXmY98AWbfPJpo5kVkpOXmRWSk5eZFZKTl5kVkpOXmRVS3clL0s6SLpN0q6RNs3kTJO3a/+GZmVVWV/KStB/wB2BzYF9WD1+2NanPLTOzpqj3m9e5wMkR8RFgRcn8acAe/RWUmVlP6k1eOwG/rDC/A9iwp5W7GzE7W/ZeSY9JWiLpt5K2KFnmEbPNrIt6k1cH6ZSx3G7ACzWs3zli9hWlM7MxGq8HvkhKgjOBa0uKTGH1iNnvAU6TNL7O2M1sAKk3eV0DXCjpX0gj/KwhaW/gItIgGVVVGTH7IOCRiLguIpaRktWYbLAN8IjZZlam3mcbzyKN2vMsIODR7PUa4Pw+xLEjJSNiR8RiSU8DO0p6mcojZk+otCFJxwHHAYwePbpXwTx3zs5d3r/WsSGwBq91PNtl2eiz/9yr7VvjuP7bR83JS9Ig0qnbp0ind7uRvrk9GBFP9jGOdYHysdk7R8Wue8Rs4HKAsWPHRqUyVrui9ajg+m8f9XzzCuBPwA4R8RTwTD/GsYg0CnapzlGxPWJ2C7lHhfY2ctj6XV7zpObkFREh6XFgI+Cpfo7jEVK7FgCS1iHdO/ZIRLwiqXPE7NuzIh4x26wJJu86sdUhdKveBvvTgIsk7SJJ9e6suxGzgRuAnSQdnC0/G3g4Ih7LVvWI2WbWRb3J6yekm1EfAJZl912tmmpYv+KI2RExDziY1Oj/CvAO4IiS9Txitpl1Ue/Vxsl92Vm1EbMj4g5gu26WecRsM+uiruQVET9qVCBmZvXodR/2kjYBhpTOi4jn+hyRmVkN6kpektYHLgEOoyxxZQb3R1BmZj2pt8H+ItJtChNI91xNBD5Heq7x8H6NzMysinpPGz8IHBkRd0l6HXggIq7N7sP6FPDTfo/QzKyCepPXCNLtCpAe0RlJumF1BvCD/gvLalG0R3fM+lO9yetpYCvgOWAWcISk+0m9QnT0c2zWAz+60xyjhq0EXsteLS9qSl6S3kZ6HOdK4G2knlO/BvyCdO/XIOCzDYnQrMVOfduCVodgFdT6zetBYNOI+AaApFuAT5JuKh0LPBkRA7Z/EB95LY/avdmg1uRV/hzju4G1IuIZ0inkgOYjr+VRuzcbeNxGMyukWpNXZFP5PDOzlqjntHGqpOXZ+2HA9yUtKS0UEQf2Z3BmRdTubVHNUmvyKn8ge2p/B2I2ULR7W1Sz1JS8IuITjQ5kIPGR1/JooH0ue92rhHXPR15rhMtOubnL+wXzF696LV02+eIDKq4/0D6XuUpekqYB7wRey2bNiYhts2UTga8Co0h92R8TEb6rv5/NOv/OLu9XdCxd9Vq6bPsz921qXGbl8nirxOSIWDebOhPXjsD3gKOAjYElwLdbGKOZtViuvnlVMQm4OSJ+ByDpi8AsScMjom2GQOvraYPZQJLHb15flTRf0t2S9snmlY+o/TSwAtim+eGZWR7kLXmdTuq1YnPSqMc3S9qaNGr2q2VlK46aLek4STMlzZw3r3wQbhvoXP/tI1enjRFxX8nbH0k6Etif6iNql2/Dw713Y6BdKq+kFfW/16V7dXk/ZMEQBjGI5xc8/4Zld//H3c0IqS3kKnlVEKS7+x8hdT8NgKStgKHAEy2Kq5AG2qXydrfOkPW6vLab3CQvSSNIg81OJ90qcTip94rPAmsCMyT9G/BH4Bzg+rw01td65PVR1/rTXlsfVHX5+R89pMv7jr+llpeOuS91WXbm1GL23p6b5EVKUOeR+gh7HXgMmBARTwBI+jRwNanr6TsA3/Vv1sZyk7wiYh6we5Xl1wDXNC8iM8uzvF1tNDOrSW6+eVn/y2ubx0+u26Omcocden+DI7Eic/KyttIOt4u0CyevAmv3S+W90YzbRWLtYCUribXzdZvhsMGDurw2SrMOEE5eBdbTpXJrjX/u9c9Wh1DRriPf8EBKQzTrfkI32JtZIfmbVwPk9bShN0YOW7/La9G8/XNXdXk/fP5CBgPPzV/YZdkDF36syZFZXzl5NUBeTxt6Y/KuE1sdgllFTl5tpFkNttZepkyZ0uV9R0fHqtfSZeXl+qrtklc7XypvVoNtO/+NrXnaLnm5Z4XGy/PfeOWQdbq8WnEN+OTlBlsrtfit+7U6BACmv3vvmsrt/bvpDY6kuNz4YWaFNOC/eZXzaYNZYw0dOrTLa6O0XfLyaUP/a9XVJsunnXfeuSn78WmjmRVSYZKXpA0l3SBpsaRnsxG0zaxNFem08VuksRo3BnYBbpH0UEQ80tKo7A2a1eZh7a0QyUvSOsDBwE4RsQj4vaSbgKOAz7c0OHuDZrV5WHsrymnjNsBrnYNxZB4ijaRtZm1IEfnv+SAb8uy6iNikZN6xwKSI2Kes7HHAcdnbbYHHK2xyFDC/zjDqXacZ++jNOkWMa35EjK9lI67/XOyjN+vUX/8RkfsJ2BVYUjbvFODmXm5vZqPXacY+2j2u3k55/X0cV337KMpp4xPAGpLeWjJvDGkkbTNrQ4VIXhGxGLgeOEfSOpL2Aj4M/G9rIzOzVilE8sqcAKwF/A34MXB89P42icubsE4z9tGbdQZSXL2V19/HcdWhEA32ZmblivTNy8xsFScvMyuktklekiZJWlRhCklnV1lvoqSZWdmXJN0q6V1Vys+WtLRsH5f1ENsRku7Lntv8W/bzCZJUZfsLJS2QdI+kT0vqsS4lTZP0iqQen9sp+z1ekXSLpDfXsU7ntFmV8qXlVpatO6mnGOvh+h+A9d/f99AUaQI+CcwFNu1m+cmkCwQHAesAawIHABdW2eZs4H11xHAK8DJwCDAcEOm+tquBodW2D6wPHAj8FfhhD/vZEngd6AAOrSGu0v0MA64Abqx1nV7URa/Xdf23Z/037YOStyn7gCwC9ulm+frZ8h4rureVkO1jMXBwX7YP7AGsJD372d16ZwN3A18HflHvfoD9gSf663fvz3Vd/+1Z/21z2lhK0gjgp8C5ETGtm2LjSEecGxoYyjhgKPDzvmwkIu4HXgD+rUqxj5GO5lcDH5C0ca3bl7Q2cDhwb1/izAvX/8Co/7ZLXlk7wlXAX4Bq43KNJD1T9VovdnNj1h7ROR3bTblR5fvI2jAWZOf/765jny8CG1ZakLXRbAH8JCIeAJ4GaukP7UZJC4BXgfcDF9a6TjbdWEvgzeT6Hzj133bJCzid1BvFxyP7vtqNvwOjJPWm26AJETGiZPp+rfuIiD0jYkS2rJ762ZzUnlHJx4FfR0Tng6/XZPN6MiGLZRgwGZguaZPqq3T53SfUsI9mc/0PkPpvq+QlaR/gTOCQiFjQQ/EZwHJgQgND6tzHh/uyEUm7kz68v6+wbC3gMGBvSXMlzQVOAsZIGlPL9iPi9Yi4ntTg2+2Vtrxz/Q+s+m+b5CVpU+D/gBMj4sGeykfEq6RGzm9JmiBpbUlrSvqgpH4ZBjr7B/oy8G1Jh0gaLmmQpF1IV7eqkrSepA+Rfq+pEfHnCsUmkD50O5B6oN0F2B64i9QO0iMlHwY2AGbVsk7euP4HYP335spAESfSBzFIV5DKp+9WWW8SMJN0VWgucAuwZ5Xys4GlZdu/oYfYJgH3A0uAecB9pD6phlTZ/kJSW8QM4DPA4G62fRtwcYX5h2W/zxo1/B4LSW1Ek3r4PWaT06uNrv+BV/9+ttHMCqltThvNbGBx8jKzQnLyMrNCcvIys0Jy8jKzQnLyMrNCcvJqAUlHS1rU1zJWTK7//uHk1QuSNpb035KelrRc0hylTur278fdXAts1YvYhkv6sqS/SFoiqUPSA5LOkDSqH+NrW67/fOjNQ6dtTdKWpH6RFgJfAB4iHQTeC3wXGN0f+4mIpaQ7nOuJbQPSYx8jgC+R7gxfALyF9CjIMXTTk4KkIRGxovcRtwfXf4406nGMgToBvwTmAOtWWDYiez0ZeJj0SMkc4Aedy7LlR5MeuziANKDuMuC3wFblZUreTyE9onEEqUuThcCNwKiSMt/JtrtZN7Gr7HGMKaQeMhcA15F62wxgbNl6QXqYmZIyE0kPAi8DHgP2a3XduP7bq/592lgHSRsC44FvRcQb2iNidU8FK4ETSV2vTCT1dHlpWfGhpKPjJ0id0g0Grs/6m+rOlqRO4T4C7EfqDfT8LLZBpA/21Ih4sdLKkX36SpxM+uCNBc6ost9KLgAuIT3oezvwc0mb17mNQnH9d9H6+m/1kaxIE+lDGMBH6lxvPKnrk0HZ+6Oz7exVUmYL0tP/7yspU37kXQasXzLvTOCp7OeNs22eVLbve1j9gPCtJfNnAzeXld2S2o+8Z5YsH0T6BnFeq+vI9d8+9e9vXvWpdlRcXUjaV9Ltkl6QtBC4HhgClHbktpLUkwAAEfEsqTfMHaps+tlIXbV0ehF4Uw/hHE46Ot5AGnG81Mwe1q1mRucPEbGS1BNCtdgHAtf/ai2vfyev+jxJOups310BSVuQuk2ZBRwKvJ3UUArpA1yq3i49/llh/c46nEdqu9iuS4GI5yPiKVL3KeUWl71fmb2u+ieVtGadMQ5krv8ccfKqQ0R0AL8CJktat3y50sAOY0kf0pMiYkZEPAFUGrtuEOk0pHPd0Vm5XnX2lh39rgU+qh7G16tiXva6acm8Xbop+87OH7J2mj3IU0d1DeD676Ll9e/kVb/PkI5MMyUdKmlbSdtJOp50helJ0t/1REn/KulIUuNtudeAb0oal/Wc+SPgEeCOPsR2BvAccK+kT0oaI2lrSQcC+5DaVLoV6fL8vcDpknaUtCdwUTfFj896/9wW+CapzeY7fYi9KFz/Sevrv9WNoEWcSEemS4FnSA2xLwK3AuOz5f9JukS+FPgNqdfKALaMksZYUt/lT2bbmA68pWQfR1PhUnlZHF3KZPPWA84DHs32v5T0T/UV4E0l5WYDp1b43bYn3ce0BPgzaTitSg22k0iNwcuAx4EPtrpeXP/tVf/uSdXqkt2k+Vdg94joS4OvFVCe6t+njWZWSE5eZlZIPm00s0LyNy8zKyQnLzMrJCcvMyskJy8zKyQnLzMrJCcvMyuk/wdNkPozk22WVwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 311.04x172.8 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "da.plotCategoricForNumTargetPairs(df,nums,cats,\"Survived\",2.4,0.9)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bunları yorumlamayı size bırakıyorum."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Checking for cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=bd827485-7a56-4fc9-a129-3e3297e97d56 style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('bd827485-7a56-4fc9-a129-3e3297e97d56').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>714.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.383838</td>\n",
       "      <td>2.308642</td>\n",
       "      <td>29.699118</td>\n",
       "      <td>0.523008</td>\n",
       "      <td>0.381594</td>\n",
       "      <td>32.204208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.486592</td>\n",
       "      <td>0.836071</td>\n",
       "      <td>14.526497</td>\n",
       "      <td>1.102743</td>\n",
       "      <td>0.806057</td>\n",
       "      <td>49.693429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.420000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>20.125000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.910400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.454200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>512.329200</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "         Survived      Pclass         Age       SibSp       Parch        Fare\n",
       "count  891.000000  891.000000  714.000000  891.000000  891.000000  891.000000\n",
       "mean     0.383838    2.308642   29.699118    0.523008    0.381594   32.204208\n",
       "std      0.486592    0.836071   14.526497    1.102743    0.806057   49.693429\n",
       "min      0.000000    1.000000    0.420000    0.000000    0.000000    0.000000\n",
       "25%      0.000000    2.000000   20.125000    0.000000    0.000000    7.910400\n",
       "50%      0.000000    3.000000   28.000000    0.000000    0.000000   14.454200\n",
       "75%      1.000000    3.000000   38.000000    1.000000    0.000000   31.000000\n",
       "max      1.000000    3.000000   80.000000    8.000000    6.000000  512.329200"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#numeric borders, check the min-max\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Age kolonunda min değer biraz garip görünüyor, bu bir bebek olabilir, bakalım yaşı 1den küçük kaç yolcu var"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=bfec6cce-fd4a-4991-8c05-469e4e5d2c5f style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('bfec6cce-fd4a-4991-8c05-469e4e5d2c5f').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>CabinGrup</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>78</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>male</td>\n",
       "      <td>0.83</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>29.0000</td>\n",
       "      <td>S</td>\n",
       "      <td>Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>305</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>151.5500</td>\n",
       "      <td>S</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>469</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>0.75</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>19.2583</td>\n",
       "      <td>C</td>\n",
       "      <td>Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>644</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>0.75</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>19.2583</td>\n",
       "      <td>C</td>\n",
       "      <td>Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>755</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>male</td>\n",
       "      <td>0.67</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>14.5000</td>\n",
       "      <td>S</td>\n",
       "      <td>Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>803</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>0.42</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.5167</td>\n",
       "      <td>C</td>\n",
       "      <td>Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>831</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>male</td>\n",
       "      <td>0.83</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>18.7500</td>\n",
       "      <td>S</td>\n",
       "      <td>Z</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "     Survived  Pclass     Sex   Age  SibSp  Parch      Fare Embarked CabinGrup\n",
       "78          1       2    male  0.83      0      2   29.0000        S         Z\n",
       "305         1       1    male  0.92      1      2  151.5500        S         C\n",
       "469         1       3  female  0.75      2      1   19.2583        C         Z\n",
       "644         1       3  female  0.75      2      1   19.2583        C         Z\n",
       "755         1       2    male  0.67      1      1   14.5000        S         Z\n",
       "803         1       3    male  0.42      0      1    8.5167        C         Z\n",
       "831         1       2    male  0.83      1      1   18.7500        S         Z"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.Age<1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bunların hepsini bebek gibi düşünebiliriz, ve belki de outlier capping yapıp yaşı 2'den küçük tüm çocukları 2 yapabilriiz."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "yukarıda low-cardinalitysi olan featureların unique valuelarına tekrar bakalım, anormal bir değer var mı, görelim. (Yok)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Diğer işlemler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#duplicate check for rows\n",
    "len(df)-len(df.duplicated(keep=False))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#duplicate check for columns\n",
    "len(set(df.columns))-len(df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=c5aa53c7-6d3c-429c-ae87-25ab5b7915f0 style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('c5aa53c7-6d3c-429c-ae87-25ab5b7915f0').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>CabinGrup</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Survived</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pclass</th>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sex</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Age</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SibSp</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Parch</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fare</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Embarked</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CabinGrup</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "           Survived  Pclass  Sex  Age  SibSp  Parch  Fare  Embarked  CabinGrup\n",
       "Survived        1.0     NaN  NaN  NaN    NaN    NaN   NaN       NaN        NaN\n",
       "Pclass          NaN     1.0  NaN  NaN    NaN    NaN   NaN       NaN        NaN\n",
       "Sex             NaN     NaN  1.0  NaN    NaN    NaN   NaN       NaN        NaN\n",
       "Age             NaN     NaN  NaN  1.0    NaN    NaN   NaN       NaN        NaN\n",
       "SibSp           NaN     NaN  NaN  NaN    1.0    NaN   NaN       NaN        NaN\n",
       "Parch           NaN     NaN  NaN  NaN    NaN    1.0   NaN       NaN        NaN\n",
       "Fare            NaN     NaN  NaN  NaN    NaN    NaN   1.0       NaN        NaN\n",
       "Embarked        NaN     NaN  NaN  NaN    NaN    NaN   NaN       1.0        NaN\n",
       "CabinGrup       NaN     NaN  NaN  NaN    NaN    NaN   NaN       NaN        1.0"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#multicollinearty check->remove one if r>0.9\n",
    "corr_results[corr_results>0.9]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "192"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#data quality to peculiar to this specific data\n",
    "len(df[df.Parch<df.SibSp]) #mesela Parch her zaman SibSp'den büyük olmalı denseydi şu kontrolü yapabilirdik"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "# any need for conversion, 100 USD-->100\n",
    "#yok"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "#check for feature extraction\n",
    "#bunu Cabin için zaten yaptık, pipelinea da eklemeyi unutmayalım"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Varsayımların kontrolü"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Çok kritik varsayımlar olmadığı ve genel kullanılan bi dataseti olduğu için pas geçiyorum."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preparing X,y and train-test splits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=72571f30-0df7-4d6c-b06e-73fcc145a6c7 style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('72571f30-0df7-4d6c-b06e-73fcc145a6c7').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>CabinGrup</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>S</td>\n",
       "      <td>Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>S</td>\n",
       "      <td>Z</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>S</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>S</td>\n",
       "      <td>Z</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "   Survived  Pclass     Sex   Age  SibSp  Parch     Fare Embarked CabinGrup\n",
       "0         0       3    male  22.0      1      0   7.2500        S         Z\n",
       "1         1       1  female  38.0      1      0  71.2833        C         C\n",
       "2         1       3  female  26.0      0      0   7.9250        S         Z\n",
       "3         1       1  female  35.0      1      0  53.1000        S         C\n",
       "4         0       3    male  35.0      0      0   8.0500        S         Z"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "X=df.iloc[:,1:]\n",
    "y=df.iloc[:,0].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "#if imbalanced add stratify=y\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=0.25, random_state=42,stratify=y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(668, 8), (223, 8), (668,), (223,)]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(map(np.shape, (X_train, X_test, y_train, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[pandas.core.frame.DataFrame,\n",
       " pandas.core.frame.DataFrame,\n",
       " numpy.ndarray,\n",
       " numpy.ndarray]"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(map(type, (X_train, X_test, y_train, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modelleme"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logistic Regression ile birlikte benzer mantıkta çalışan birkaç classfiera daha bakacağız."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression,RidgeClassifier,SGDClassifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Diğer ikisini kısaca tanıyalım:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Lasso**\n",
    "\n",
    "LinReg notebookunda detay bilgi var, <a href=\"https://machinelearningmastery.com/lasso-regression-with-python/\">şuraya</a> da bakabilirsiniz. Burda belirtildiği gibi Lasso, en küçük kareler problemini L1 cezasıyla(penaltı) optimize eder. Tanım olarak, Lasso ile bir lojistik fonksiyonu optimize edemezsiniz. L1 cezalı bir lojistik fonksiyonu optimize etmek istiyorsanız, L1 parametreli LogReg modeli kullanabilirsiniz. Yani özetle, LinReg'de olduğunun aksine Lasso sınıfnı kullanmayacağız, bunun için penalty parametresini kullanacağız."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**RidgeClassifer**\n",
    "\n",
    "Bu da L2 penaltısı kullanır, LogReg'den daha hızlı."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**SGDClassifier**\n",
    "\n",
    "SGD, ayrı bi tahminleme algoritması değil, bir optimizasyon yöntemidir. Lineer regresyon notebookunda bunu incelemiştik. SGDClassifer, sklearn içinde her ne kadar ayrı bir algoritmaymış gibi tutulmuş olsa da aslında bu optimizasyon yöntemi olarak SGD uygulayan lineer bir classifierdır. Burda loss=\"log_loss dediğimizde aslında optimizer olarak OLS değil de SGD kullanan bir LogReg eğitmiş oluruz.(log-loss yerine hinge kullanılırsa da bu sefer SVM eğitilmiş olur)\n",
    "\n",
    "Özellikle sample sayısı çokken, yani large data sözkonusu iken işe yarar, hem RAM hem cpu açısından daha verimli.\n",
    "\n",
    "- https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.SGDClassifier.html\n",
    "- https://scikit-learn.org/stable/modules/sgd.html#sgd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### hyperparametreler"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "LogReg'in birçok hyperparametresi var ancak eğitimlerde çoğunlukla 1-2 tanesi üzerinde durulup geçilir. Ben öyle yapmanızı tavsiye etmiyorum. Buraya da copy paste yapmanın bi anlamı yok, dokümantasyonundan lütfen her parametrenin ne olduğuna, hangisinin neyle birlikte kullanılması gerektiğine bakın. Bunları doğru yapmazsanız ya eğitimlerde yüzeysel anlatıldığı gibi basit bir gridsearch yapmış olursunuz ve bu da muhtemelen yetersiz kalır, çünkü tüm olası seçenekleri denememiş olursunuz, ya da herşeyi birlikte denemeye çalışıp hatalar alırsınız, çünkü her parametrenin her değeri birbiriyle uyumlu değil. O yüzden birbiriyle uyumlu olacak geniş bir parametre uzayını denemek için bunlara hakim olmanız gerekir."
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)\n",
    "\n",
    "https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Solverlar(oppitmizerlar) hakkında detaylı bilgiyi <a href=\"https://stackoverflow.com/questions/38640109/logistic-regression-python-solvers-definitions\">buradan</a> ve <a href=\"https://towardsdatascience.com/dont-sweat-the-solver-stuff-aea7cddc3451\">buradan</a> elde edebilrsiniz."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Pipeline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Şimdi yavaştan pipelinımızı kurmaya başlayalım. Öncelikle outlierhandlerımız olan custom class tanımımızı yapalım."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "\n",
    "def logTransformer(df,col_name):    \n",
    "    temp=df.copy() #her defasında orjinal X_train bozulmasın diye, yoksa gridi 2.kez çalıştırdığımda hata alıyoruz\n",
    "    temp[col_name] = np.log10(temp[col_name]+0.01)\n",
    "    return temp\n",
    "   \n",
    "class OutlierHandler(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, featureindices): #if only specific columns to be processed\n",
    "        self.featureindices = featureindices\n",
    "  \n",
    "    def fit(self, X:np.array, y = None):\n",
    "        Q1s = np.quantile(X[:,self.featureindices],0.1,axis=0)\n",
    "        Q3s = np.quantile(X[:,self.featureindices],0.9,axis=0)\n",
    "        IQRs = Q3s-Q1s\n",
    "        self.top=(Q3s + 1.5 * IQRs)\n",
    "        self.bottom=(Q1s - 1.5 * IQRs)\n",
    "        return self \n",
    "    \n",
    "    def transform(self, X:np.array, y = None ):\n",
    "        X[:,self.featureindices]=np.where(X[:,self.featureindices]>self.top,self.top,X[:,self.featureindices])\n",
    "        X[:,self.featureindices]=np.where(X[:,self.featureindices]<self.bottom,self.bottom,X[:,self.featureindices])\n",
    "        return X"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "parametergrid içinde kullanılacak bazı variableları tanımlayalım."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "c_range=np.logspace(4,-5, num=10) #ilk run'da ideal noktayı bulacağız. iknci run'da idealin etrafında daraltma yapacağız\n",
    "weights = np.arange(2,11,2)\n",
    "min_res=50\n",
    "fact=3\n",
    "mi=4000 #max_iter for LogReg"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ve tüm pipelınımız da şöyle olur"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "n_iterations: 3\n",
      "n_required_iterations: 3\n",
      "n_possible_iterations: 3\n",
      "min_resources_: 50\n",
      "max_resources_: 668\n",
      "aggressive_elimination: False\n",
      "factor: 3\n",
      "----------\n",
      "iter: 0\n",
      "n_candidates: 13\n",
      "n_resources: 50\n",
      "Fitting 50 folds for each of 13 candidates, totalling 650 fits\n",
      "----------\n",
      "iter: 1\n",
      "n_candidates: 5\n",
      "n_resources: 150\n",
      "Fitting 50 folds for each of 5 candidates, totalling 250 fits\n",
      "----------\n",
      "iter: 2\n",
      "n_candidates: 2\n",
      "n_resources: 450\n",
      "Fitting 50 folds for each of 2 candidates, totalling 100 fits\n",
      "Wall time: 1min 49s\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>HalvingRandomSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                      error_score=&#x27;raise&#x27;,\n",
       "                      estimator=Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                                                 FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                                                     kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                                                (&#x27;ct&#x27;,\n",
       "                                                 ColumnTransformer(n_jobs=-1,\n",
       "                                                                   remainder=&#x27;passthrough&#x27;,\n",
       "                                                                   transformers=[(&#x27;nominals&#x27;,\n",
       "                                                                                  Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                                                   OneHotEncod...\n",
       "                                            &#x27;clf__alpha&#x27;: array([1.e+04, 1.e+03, 1.e+02, 1.e+01, 1.e+00, 1.e-01, 1.e-02, 1.e-03,\n",
       "       1.e-04, 1.e-05]),\n",
       "                                            &#x27;clf__class_weight&#x27;: [{0: 1, 1: 2},\n",
       "                                                                  {0: 1, 1: 4},\n",
       "                                                                  {0: 1, 1: 6},\n",
       "                                                                  {0: 1, 1: 8},\n",
       "                                                                  {0: 1, 1: 10},\n",
       "                                                                  &#x27;balanced&#x27;],\n",
       "                                            &#x27;clf__solver&#x27;: [&#x27;svd&#x27;, &#x27;cholesky&#x27;,\n",
       "                                                            &#x27;lsqr&#x27;, &#x27;sparse_cg&#x27;,\n",
       "                                                            &#x27;sag&#x27;, &#x27;saga&#x27;],\n",
       "                                            &#x27;clf__tol&#x27;: [0.001, 0.0001],\n",
       "                                            &#x27;ct__numerics__ouh&#x27;: [OutlierHandler(featureindices=[0,\n",
       "                                                                                                 3]),\n",
       "                                                                  None],\n",
       "                                            &#x27;ct__numerics__scl&#x27;: [StandardScaler(),\n",
       "                                                                  MinMaxScaler()]}],\n",
       "                      scoring=&#x27;accuracy&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">HalvingRandomSearchCV</label><div class=\"sk-toggleable__content\"><pre>HalvingRandomSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                      error_score=&#x27;raise&#x27;,\n",
       "                      estimator=Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                                                 FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                                                     kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                                                (&#x27;ct&#x27;,\n",
       "                                                 ColumnTransformer(n_jobs=-1,\n",
       "                                                                   remainder=&#x27;passthrough&#x27;,\n",
       "                                                                   transformers=[(&#x27;nominals&#x27;,\n",
       "                                                                                  Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                                                   OneHotEncod...\n",
       "                                            &#x27;clf__alpha&#x27;: array([1.e+04, 1.e+03, 1.e+02, 1.e+01, 1.e+00, 1.e-01, 1.e-02, 1.e-03,\n",
       "       1.e-04, 1.e-05]),\n",
       "                                            &#x27;clf__class_weight&#x27;: [{0: 1, 1: 2},\n",
       "                                                                  {0: 1, 1: 4},\n",
       "                                                                  {0: 1, 1: 6},\n",
       "                                                                  {0: 1, 1: 8},\n",
       "                                                                  {0: 1, 1: 10},\n",
       "                                                                  &#x27;balanced&#x27;],\n",
       "                                            &#x27;clf__solver&#x27;: [&#x27;svd&#x27;, &#x27;cholesky&#x27;,\n",
       "                                                            &#x27;lsqr&#x27;, &#x27;sparse_cg&#x27;,\n",
       "                                                            &#x27;sag&#x27;, &#x27;saga&#x27;],\n",
       "                                            &#x27;clf__tol&#x27;: [0.001, 0.0001],\n",
       "                                            &#x27;ct__numerics__ouh&#x27;: [OutlierHandler(featureindices=[0,\n",
       "                                                                                                 3]),\n",
       "                                                                  None],\n",
       "                                            &#x27;ct__numerics__scl&#x27;: [StandardScaler(),\n",
       "                                                                  MinMaxScaler()]}],\n",
       "                      scoring=&#x27;accuracy&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                 FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                     kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                (&#x27;ct&#x27;,\n",
       "                 ColumnTransformer(n_jobs=-1, remainder=&#x27;passthrough&#x27;,\n",
       "                                   transformers=[(&#x27;nominals&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                   OneHotEncoder(drop=&#x27;first&#x27;,\n",
       "                                                                                 handle_unknown=&#x27;ignore&#x27;))]),\n",
       "                                                  [1, 6, 7]),\n",
       "                                                 (&#x27;numerics&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;imp&#x27;,\n",
       "                                                                   SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                                                                  (&#x27;ouh&#x27;,\n",
       "                                                                   DummyTransformer()),\n",
       "                                                                  (&#x27;scl&#x27;,\n",
       "                                                                   DummyTransformer())]),\n",
       "                                                  [2, 3, 4, 5])])),\n",
       "                (&#x27;fs&#x27;,\n",
       "                 SelectKBest(score_func=&lt;function mutual_info_classif at 0x000001BC26E09B80&gt;)),\n",
       "                (&#x27;clf&#x27;, DummyEstimator())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                    kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})</pre></div></div></div><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">ct: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(n_jobs=-1, remainder=&#x27;passthrough&#x27;,\n",
       "                  transformers=[(&#x27;nominals&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                  OneHotEncoder(drop=&#x27;first&#x27;,\n",
       "                                                                handle_unknown=&#x27;ignore&#x27;))]),\n",
       "                                 [1, 6, 7]),\n",
       "                                (&#x27;numerics&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;imp&#x27;,\n",
       "                                                  SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                                                 (&#x27;ouh&#x27;, DummyTransformer()),\n",
       "                                                 (&#x27;scl&#x27;, DummyTransformer())]),\n",
       "                                 [2, 3, 4, 5])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">nominals</label><div class=\"sk-toggleable__content\"><pre>[1, 6, 7]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" ><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder(drop=&#x27;first&#x27;, handle_unknown=&#x27;ignore&#x27;)</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-7\" type=\"checkbox\" ><label for=\"sk-estimator-id-7\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">numerics</label><div class=\"sk-toggleable__content\"><pre>[2, 3, 4, 5]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-8\" type=\"checkbox\" ><label for=\"sk-estimator-id-8\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;median&#x27;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-9\" type=\"checkbox\" ><label for=\"sk-estimator-id-9\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyTransformer</label><div class=\"sk-toggleable__content\"><pre>DummyTransformer()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-10\" type=\"checkbox\" ><label for=\"sk-estimator-id-10\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyTransformer</label><div class=\"sk-toggleable__content\"><pre>DummyTransformer()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-11\" type=\"checkbox\" ><label for=\"sk-estimator-id-11\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">remainder</label><div class=\"sk-toggleable__content\"><pre></pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-12\" type=\"checkbox\" ><label for=\"sk-estimator-id-12\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">passthrough</label><div class=\"sk-toggleable__content\"><pre>passthrough</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-13\" type=\"checkbox\" ><label for=\"sk-estimator-id-13\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SelectKBest</label><div class=\"sk-toggleable__content\"><pre>SelectKBest(score_func=&lt;function mutual_info_classif at 0x000001BC26E09B80&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-14\" type=\"checkbox\" ><label for=\"sk-estimator-id-14\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyEstimator</label><div class=\"sk-toggleable__content\"><pre>DummyEstimator()</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "HalvingRandomSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                      error_score='raise',\n",
       "                      estimator=Pipeline(steps=[('log',\n",
       "                                                 FunctionTransformer(func=<function logTransformer at 0x000001BC2B98CE50>,\n",
       "                                                                     kw_args={'col_name': 'Fare'})),\n",
       "                                                ('ct',\n",
       "                                                 ColumnTransformer(n_jobs=-1,\n",
       "                                                                   remainder='passthrough',\n",
       "                                                                   transformers=[('nominals',\n",
       "                                                                                  Pipeline(steps=[('ohe',\n",
       "                                                                                                   OneHotEncod...\n",
       "                                            'clf__alpha': array([1.e+04, 1.e+03, 1.e+02, 1.e+01, 1.e+00, 1.e-01, 1.e-02, 1.e-03,\n",
       "       1.e-04, 1.e-05]),\n",
       "                                            'clf__class_weight': [{0: 1, 1: 2},\n",
       "                                                                  {0: 1, 1: 4},\n",
       "                                                                  {0: 1, 1: 6},\n",
       "                                                                  {0: 1, 1: 8},\n",
       "                                                                  {0: 1, 1: 10},\n",
       "                                                                  'balanced'],\n",
       "                                            'clf__solver': ['svd', 'cholesky',\n",
       "                                                            'lsqr', 'sparse_cg',\n",
       "                                                            'sag', 'saga'],\n",
       "                                            'clf__tol': [0.001, 0.0001],\n",
       "                                            'ct__numerics__ouh': [OutlierHandler(featureindices=[0,\n",
       "                                                                                                 3]),\n",
       "                                                                  None],\n",
       "                                            'ct__numerics__scl': [StandardScaler(),\n",
       "                                                                  MinMaxScaler()]}],\n",
       "                      scoring='accuracy', verbose=1)"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "from sklearn.experimental import enable_halving_search_cv\n",
    "from sklearn.model_selection import HalvingRandomSearchCV\n",
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "\n",
    "np.random.seed(42) #RandomizedSearchCV için\n",
    "\n",
    "class DummyTransformer(TransformerMixin,BaseEstimator):\n",
    "    def fit(self,X,y=None): pass\n",
    "    def transform(self,X,y=None): pass\n",
    "    \n",
    "class DummyEstimator(BaseEstimator):\n",
    "    def fit(self,X,y=None): pass\n",
    "    def score(self,X,y=None): pass    \n",
    "    \n",
    "cat_pipe=Pipeline([ \n",
    "                   (\"ohe\", OneHotEncoder(drop=\"first\",handle_unknown='ignore'))     \n",
    "                  ])\n",
    "\n",
    "num_pipe=Pipeline([\n",
    "                   (\"imp\", SimpleImputer(strategy=\"median\")),     \n",
    "                   (\"ouh\", DummyTransformer()), \n",
    "                   (\"scl\", DummyTransformer())\n",
    "                  ])\n",
    "\n",
    "\n",
    "coltrans = ColumnTransformer([\n",
    "                                ('nominals',  cat_pipe, [e for e,v in enumerate(X.columns) if v in noms]),\n",
    "                                ('numerics',  num_pipe, [e for e,v in enumerate(X.columns) if v in nums])\n",
    "                                 ],n_jobs=-1,remainder=\"passthrough\") #ordinaller\n",
    "\n",
    "\n",
    "pipe = Pipeline(steps=[('log', FunctionTransformer(logTransformer,kw_args=dict(col_name=\"Fare\"))),\n",
    "                       ('ct', coltrans),\n",
    "                       ('fs', SelectKBest(score_func=mutual_info_classif,k=10)),\n",
    "                       ('clf', DummyEstimator()) \n",
    "                       ])\n",
    "\n",
    "params = [          \n",
    "          {\n",
    "           'clf'         : [LogisticRegression(max_iter=mi,random_state=42)],\n",
    "           'clf__C'      : c_range, \n",
    "           'clf__tol'      :[0.001, 0.0001], \n",
    "           'clf__penalty': ['l2'], #none koymadım, çünkü zaten yüksek C değeri de veriyorum ki bu zaten regularizasyon uygulama demekle aynı şey\n",
    "           'clf__solver' : ['newton-cg', 'lbfgs','sag','saga'],\n",
    "           'clf__class_weight': [{0:1, 1:x} for x in weights] + ['balanced'],\n",
    "           'ct__numerics__ouh': [OutlierHandler(featureindices=[0,3]),None], #Age ve Fare için, her ne kadar Fare'de LogTrans yapmış olsak da\n",
    "           'ct__numerics__scl': [StandardScaler(),MinMaxScaler()]\n",
    "          } ,\n",
    "          {\n",
    "           'clf'         : [LogisticRegression(max_iter=mi,random_state=42)],\n",
    "           'clf__C'      : c_range, \n",
    "           'clf__tol'      :[0.000, 0.0001],               \n",
    "           'clf__penalty': ['l1'], \n",
    "           'clf__solver' : ['liblinear','saga'],\n",
    "           'clf__class_weight': [{0:1, 1:x} for x in weights] + ['balanced'],                \n",
    "           'ct__numerics__ouh': [OutlierHandler(featureindices=[0,3]),None], #Age ve Fare için, her ne kadar Fare'de LogTrans yapmış olsak da\n",
    "           'ct__numerics__scl': [StandardScaler(),MinMaxScaler()]\n",
    "          } ,    \n",
    "          {\n",
    "           'clf'         : [LogisticRegression(max_iter=mi,random_state=42)],\n",
    "           'clf__C'      : c_range, \n",
    "           'clf__tol'      :[0.001, 0.0001],                             \n",
    "           'clf__penalty': ['elasticnet'], \n",
    "           'clf__l1_ratio':[0.15, 0.5, 1],\n",
    "           'clf__solver' : ['saga'],\n",
    "           'clf__class_weight': [{0:1, 1:x} for x in weights] + ['balanced'],              \n",
    "           'ct__numerics__ouh': [OutlierHandler(featureindices=[0,3]),None], #Age ve Fare için, her ne kadar Fare'de LogTrans yapmış olsak da\n",
    "           'ct__numerics__scl': [StandardScaler(),MinMaxScaler()]\n",
    "          } ,        \n",
    "          {\n",
    "           'clf'               : [SGDClassifier(loss='log_loss',max_iter=mi,random_state=42)],\n",
    "           'clf__alpha'        : c_range, \n",
    "           'clf__tol'          : [0.001, 0.0001],                             \n",
    "           'clf__penalty'      : ['l1','l2','elasticnet'],    \n",
    "           'clf__l1_ratio'     : [0.15, 0.5, 1],   \n",
    "           'clf__learning_rate': ['constant','optimal','invscaling','adaptive'],\n",
    "           'clf__eta0'          : [0.01, 0.001, 0.0001],    \n",
    "           'clf__early_stopping':[True,False], \n",
    "           'clf__class_weight': [{0:1, 1:x} for x in weights] + ['balanced'],                \n",
    "           'ct__numerics__ouh': [OutlierHandler(featureindices=[0,3]),None], #Age ve Fare için, her ne kadar Fare'de LogTrans yapmış olsak da\n",
    "           'ct__numerics__scl' : [StandardScaler(),MinMaxScaler()]\n",
    "          } ,  \n",
    "          {\n",
    "           'clf'              : [RidgeClassifier(random_state=42)],\n",
    "           'clf__alpha'       : c_range, \n",
    "           'clf__tol'         :[0.001, 0.0001],            \n",
    "           'clf__solver'      : ['svd', 'cholesky', 'lsqr', 'sparse_cg', 'sag', 'saga'],\n",
    "           'clf__class_weight': [{0:1, 1:x} for x in weights] + ['balanced'],                \n",
    "           'ct__numerics__ouh': [OutlierHandler(featureindices=[0,3]),None], #Age ve Fare için, her ne kadar Fare'de LogTrans yapmış olsak da\n",
    "           'ct__numerics__scl': [StandardScaler(),MinMaxScaler()]\n",
    "          }    \n",
    "         ]         \n",
    "\n",
    "mycv = RepeatedKFold(n_splits=5, n_repeats=10, random_state=1)\n",
    "hrs1 = HalvingRandomSearchCV(estimator = pipe, param_distributions = params, cv = mycv, n_jobs=-1, verbose = 1, \n",
    "                           scoring = 'accuracy',error_score='raise',min_resources=min_res,factor=fact) \n",
    "hrs1.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=ae0e3744-a0d2-4800-aebe-33f8d1e91287 style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('ae0e3744-a0d2-4800-aebe-33f8d1e91287').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>param_ct__numerics__scl</th>\n",
       "      <th>param_ct__numerics__ouh</th>\n",
       "      <th>param_clf__tol</th>\n",
       "      <th>param_clf__penalty</th>\n",
       "      <th>param_clf__learning_rate</th>\n",
       "      <th>param_clf__l1_ratio</th>\n",
       "      <th>param_clf__eta0</th>\n",
       "      <th>param_clf__early_stopping</th>\n",
       "      <th>param_clf__class_weight</th>\n",
       "      <th>param_clf__alpha</th>\n",
       "      <th>param_clf</th>\n",
       "      <th>param_clf__solver</th>\n",
       "      <th>param_clf__C</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>StandardScaler()</td>\n",
       "      <td>OutlierHandler(featureindices=[0, 3])</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.001</td>\n",
       "      <td>RidgeClassifier(alpha=0.001, class_weight='balanced', random_state=42,\\n                solver='cholesky', tol=0.0001)</td>\n",
       "      <td>cholesky</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.791129</td>\n",
       "      <td>0.039551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>StandardScaler()</td>\n",
       "      <td>OutlierHandler(featureindices=[0, 3])</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.001</td>\n",
       "      <td>RidgeClassifier(alpha=0.001, class_weight='balanced', random_state=42,\\n                solver='cholesky', tol=0.0001)</td>\n",
       "      <td>cholesky</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.772713</td>\n",
       "      <td>0.075316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>StandardScaler()</td>\n",
       "      <td>OutlierHandler(featureindices=[0, 3])</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>elasticnet</td>\n",
       "      <td>constant</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>False</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.01</td>\n",
       "      <td>SGDClassifier(loss='log_loss', max_iter=4000, random_state=42)</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.760350</td>\n",
       "      <td>0.047239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>StandardScaler()</td>\n",
       "      <td>OutlierHandler(featureindices=[0, 3])</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.001</td>\n",
       "      <td>RidgeClassifier(alpha=0.001, class_weight='balanced', random_state=42,\\n                solver='cholesky', tol=0.0001)</td>\n",
       "      <td>cholesky</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.740222</td>\n",
       "      <td>0.135790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>StandardScaler()</td>\n",
       "      <td>OutlierHandler(featureindices=[0, 3])</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>elasticnet</td>\n",
       "      <td>constant</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>False</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.01</td>\n",
       "      <td>SGDClassifier(loss='log_loss', max_iter=4000, random_state=42)</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.733218</td>\n",
       "      <td>0.076966</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "   param_ct__numerics__scl                param_ct__numerics__ouh  \\\n",
       "19        StandardScaler()  OutlierHandler(featureindices=[0, 3])   \n",
       "17        StandardScaler()  OutlierHandler(featureindices=[0, 3])   \n",
       "18        StandardScaler()  OutlierHandler(featureindices=[0, 3])   \n",
       "3         StandardScaler()  OutlierHandler(featureindices=[0, 3])   \n",
       "16        StandardScaler()  OutlierHandler(featureindices=[0, 3])   \n",
       "\n",
       "   param_clf__tol param_clf__penalty param_clf__learning_rate  \\\n",
       "19         0.0001                NaN                      NaN   \n",
       "17         0.0001                NaN                      NaN   \n",
       "18         0.0001         elasticnet                 constant   \n",
       "3          0.0001                NaN                      NaN   \n",
       "16         0.0001         elasticnet                 constant   \n",
       "\n",
       "   param_clf__l1_ratio param_clf__eta0 param_clf__early_stopping  \\\n",
       "19                 NaN             NaN                       NaN   \n",
       "17                 NaN             NaN                       NaN   \n",
       "18                   1          0.0001                     False   \n",
       "3                  NaN             NaN                       NaN   \n",
       "16                   1          0.0001                     False   \n",
       "\n",
       "   param_clf__class_weight param_clf__alpha  \\\n",
       "19                balanced            0.001   \n",
       "17                balanced            0.001   \n",
       "18                balanced             0.01   \n",
       "3                 balanced            0.001   \n",
       "16                balanced             0.01   \n",
       "\n",
       "                                            param_clf param_clf__solver  \\\n",
       "19  RidgeClassifier(alpha=0.001, class_weight='bal...          cholesky   \n",
       "17  RidgeClassifier(alpha=0.001, class_weight='bal...          cholesky   \n",
       "18  SGDClassifier(loss='log_loss', max_iter=4000, ...               NaN   \n",
       "3   RidgeClassifier(alpha=0.001, class_weight='bal...          cholesky   \n",
       "16  SGDClassifier(loss='log_loss', max_iter=4000, ...               NaN   \n",
       "\n",
       "   param_clf__C  mean_test_score  std_test_score  \n",
       "19          NaN         0.791129        0.039551  \n",
       "17          NaN         0.772713        0.075316  \n",
       "18          NaN         0.760350        0.047239  \n",
       "3           NaN         0.740222        0.135790  \n",
       "16          NaN         0.733218        0.076966  "
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ml.gridsearch_to_df(hrs1,5)    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=8d4ccaec-f408-4822-8f96-f7a36196a0d5 style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('8d4ccaec-f408-4822-8f96-f7a36196a0d5').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>MAX of mean_test_score</th>\n",
       "      <th>MIN of mean_fit_time</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param_clf</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>RidgeClassifier</th>\n",
       "      <td>0.791129</td>\n",
       "      <td>0.370013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SGDClassifier</th>\n",
       "      <td>0.760350</td>\n",
       "      <td>0.311823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LogisticRegression</th>\n",
       "      <td>0.646000</td>\n",
       "      <td>0.358652</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "                    MAX of mean_test_score  MIN of mean_fit_time\n",
       "param_clf                                                       \n",
       "RidgeClassifier                   0.791129              0.370013\n",
       "SGDClassifier                     0.760350              0.311823\n",
       "LogisticRegression                0.646000              0.358652"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ml.compareEstimatorsInGridSearch(hrs1,tableorplot='table')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ml.compareEstimatorsInGridSearch(hrs1,tableorplot='plot',figsize=(4,4))        "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Normal veya halving gridsearch yapsaydık 5664000 adet model çalışacaktı, bunu local makinede yapmak imkansız gibi. Peki halvingsiz normal randomsearch yapsak, ne kadar sürerdi ona bakalım, bi de skorumuz çok farkediyor mu, yani halvingrandom çalışıtırmak değmiş mi görelim,"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 50 folds for each of 100 candidates, totalling 5000 fits\n",
      "Wall time: 9min 42s\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;background-color: white;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomizedSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                   error_score=&#x27;raise&#x27;,\n",
       "                   estimator=Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                                              FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                                                  kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                                             (&#x27;ct&#x27;,\n",
       "                                              ColumnTransformer(n_jobs=-1,\n",
       "                                                                remainder=&#x27;passthrough&#x27;,\n",
       "                                                                transformers=[(&#x27;nominals&#x27;,\n",
       "                                                                               Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                                                OneHotEncoder(...\n",
       "                                         &#x27;clf__alpha&#x27;: array([1.e+04, 1.e+03, 1.e+02, 1.e+01, 1.e+00, 1.e-01, 1.e-02, 1.e-03,\n",
       "       1.e-04, 1.e-05]),\n",
       "                                         &#x27;clf__class_weight&#x27;: [{0: 1, 1: 2},\n",
       "                                                               {0: 1, 1: 4},\n",
       "                                                               {0: 1, 1: 6},\n",
       "                                                               {0: 1, 1: 8},\n",
       "                                                               {0: 1, 1: 10},\n",
       "                                                               &#x27;balanced&#x27;],\n",
       "                                         &#x27;clf__solver&#x27;: [&#x27;svd&#x27;, &#x27;cholesky&#x27;,\n",
       "                                                         &#x27;lsqr&#x27;, &#x27;sparse_cg&#x27;,\n",
       "                                                         &#x27;sag&#x27;, &#x27;saga&#x27;],\n",
       "                                         &#x27;clf__tol&#x27;: [0.001, 0.0001],\n",
       "                                         &#x27;ct__numerics__ouh&#x27;: [OutlierHandler(featureindices=[0,\n",
       "                                                                                              3]),\n",
       "                                                               None],\n",
       "                                         &#x27;ct__numerics__scl&#x27;: [StandardScaler(),\n",
       "                                                               MinMaxScaler()]}],\n",
       "                   scoring=&#x27;accuracy&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-15\" type=\"checkbox\" ><label for=\"sk-estimator-id-15\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomizedSearchCV</label><div class=\"sk-toggleable__content\"><pre>RandomizedSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                   error_score=&#x27;raise&#x27;,\n",
       "                   estimator=Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                                              FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                                                  kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                                             (&#x27;ct&#x27;,\n",
       "                                              ColumnTransformer(n_jobs=-1,\n",
       "                                                                remainder=&#x27;passthrough&#x27;,\n",
       "                                                                transformers=[(&#x27;nominals&#x27;,\n",
       "                                                                               Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                                                OneHotEncoder(...\n",
       "                                         &#x27;clf__alpha&#x27;: array([1.e+04, 1.e+03, 1.e+02, 1.e+01, 1.e+00, 1.e-01, 1.e-02, 1.e-03,\n",
       "       1.e-04, 1.e-05]),\n",
       "                                         &#x27;clf__class_weight&#x27;: [{0: 1, 1: 2},\n",
       "                                                               {0: 1, 1: 4},\n",
       "                                                               {0: 1, 1: 6},\n",
       "                                                               {0: 1, 1: 8},\n",
       "                                                               {0: 1, 1: 10},\n",
       "                                                               &#x27;balanced&#x27;],\n",
       "                                         &#x27;clf__solver&#x27;: [&#x27;svd&#x27;, &#x27;cholesky&#x27;,\n",
       "                                                         &#x27;lsqr&#x27;, &#x27;sparse_cg&#x27;,\n",
       "                                                         &#x27;sag&#x27;, &#x27;saga&#x27;],\n",
       "                                         &#x27;clf__tol&#x27;: [0.001, 0.0001],\n",
       "                                         &#x27;ct__numerics__ouh&#x27;: [OutlierHandler(featureindices=[0,\n",
       "                                                                                              3]),\n",
       "                                                               None],\n",
       "                                         &#x27;ct__numerics__scl&#x27;: [StandardScaler(),\n",
       "                                                               MinMaxScaler()]}],\n",
       "                   scoring=&#x27;accuracy&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-16\" type=\"checkbox\" ><label for=\"sk-estimator-id-16\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                 FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                     kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                (&#x27;ct&#x27;,\n",
       "                 ColumnTransformer(n_jobs=-1, remainder=&#x27;passthrough&#x27;,\n",
       "                                   transformers=[(&#x27;nominals&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                   OneHotEncoder(drop=&#x27;first&#x27;,\n",
       "                                                                                 handle_unknown=&#x27;ignore&#x27;))]),\n",
       "                                                  [1, 6, 7]),\n",
       "                                                 (&#x27;numerics&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;imp&#x27;,\n",
       "                                                                   SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                                                                  (&#x27;ouh&#x27;,\n",
       "                                                                   DummyTransformer()),\n",
       "                                                                  (&#x27;scl&#x27;,\n",
       "                                                                   DummyTransformer())]),\n",
       "                                                  [2, 3, 4, 5])])),\n",
       "                (&#x27;fs&#x27;,\n",
       "                 SelectKBest(score_func=&lt;function mutual_info_classif at 0x000001BC26E09B80&gt;)),\n",
       "                (&#x27;clf&#x27;, DummyEstimator())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-17\" type=\"checkbox\" ><label for=\"sk-estimator-id-17\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                    kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})</pre></div></div></div><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-18\" type=\"checkbox\" ><label for=\"sk-estimator-id-18\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">ct: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(n_jobs=-1, remainder=&#x27;passthrough&#x27;,\n",
       "                  transformers=[(&#x27;nominals&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                  OneHotEncoder(drop=&#x27;first&#x27;,\n",
       "                                                                handle_unknown=&#x27;ignore&#x27;))]),\n",
       "                                 [1, 6, 7]),\n",
       "                                (&#x27;numerics&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;imp&#x27;,\n",
       "                                                  SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                                                 (&#x27;ouh&#x27;, DummyTransformer()),\n",
       "                                                 (&#x27;scl&#x27;, DummyTransformer())]),\n",
       "                                 [2, 3, 4, 5])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-19\" type=\"checkbox\" ><label for=\"sk-estimator-id-19\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">nominals</label><div class=\"sk-toggleable__content\"><pre>[1, 6, 7]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-20\" type=\"checkbox\" ><label for=\"sk-estimator-id-20\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder(drop=&#x27;first&#x27;, handle_unknown=&#x27;ignore&#x27;)</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-21\" type=\"checkbox\" ><label for=\"sk-estimator-id-21\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">numerics</label><div class=\"sk-toggleable__content\"><pre>[2, 3, 4, 5]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-22\" type=\"checkbox\" ><label for=\"sk-estimator-id-22\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;median&#x27;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-23\" type=\"checkbox\" ><label for=\"sk-estimator-id-23\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyTransformer</label><div class=\"sk-toggleable__content\"><pre>DummyTransformer()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-24\" type=\"checkbox\" ><label for=\"sk-estimator-id-24\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyTransformer</label><div class=\"sk-toggleable__content\"><pre>DummyTransformer()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-25\" type=\"checkbox\" ><label for=\"sk-estimator-id-25\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">remainder</label><div class=\"sk-toggleable__content\"><pre></pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-26\" type=\"checkbox\" ><label for=\"sk-estimator-id-26\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">passthrough</label><div class=\"sk-toggleable__content\"><pre>passthrough</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-27\" type=\"checkbox\" ><label for=\"sk-estimator-id-27\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SelectKBest</label><div class=\"sk-toggleable__content\"><pre>SelectKBest(score_func=&lt;function mutual_info_classif at 0x000001BC26E09B80&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-28\" type=\"checkbox\" ><label for=\"sk-estimator-id-28\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyEstimator</label><div class=\"sk-toggleable__content\"><pre>DummyEstimator()</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomizedSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                   error_score='raise',\n",
       "                   estimator=Pipeline(steps=[('log',\n",
       "                                              FunctionTransformer(func=<function logTransformer at 0x000001BC2B98CE50>,\n",
       "                                                                  kw_args={'col_name': 'Fare'})),\n",
       "                                             ('ct',\n",
       "                                              ColumnTransformer(n_jobs=-1,\n",
       "                                                                remainder='passthrough',\n",
       "                                                                transformers=[('nominals',\n",
       "                                                                               Pipeline(steps=[('ohe',\n",
       "                                                                                                OneHotEncoder(...\n",
       "                                         'clf__alpha': array([1.e+04, 1.e+03, 1.e+02, 1.e+01, 1.e+00, 1.e-01, 1.e-02, 1.e-03,\n",
       "       1.e-04, 1.e-05]),\n",
       "                                         'clf__class_weight': [{0: 1, 1: 2},\n",
       "                                                               {0: 1, 1: 4},\n",
       "                                                               {0: 1, 1: 6},\n",
       "                                                               {0: 1, 1: 8},\n",
       "                                                               {0: 1, 1: 10},\n",
       "                                                               'balanced'],\n",
       "                                         'clf__solver': ['svd', 'cholesky',\n",
       "                                                         'lsqr', 'sparse_cg',\n",
       "                                                         'sag', 'saga'],\n",
       "                                         'clf__tol': [0.001, 0.0001],\n",
       "                                         'ct__numerics__ouh': [OutlierHandler(featureindices=[0,\n",
       "                                                                                              3]),\n",
       "                                                               None],\n",
       "                                         'ct__numerics__scl': [StandardScaler(),\n",
       "                                                               MinMaxScaler()]}],\n",
       "                   scoring='accuracy', verbose=1)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "rs1 = RandomizedSearchCV(estimator=pipe, param_distributions = params, cv = mycv, n_jobs=-1, verbose = 1, \n",
    "                          scoring = 'accuracy',error_score='raise',n_iter=100)\n",
    "rs1.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=747c850f-bb5a-4b91-897a-ca4171f6b173 style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('747c850f-bb5a-4b91-897a-ca4171f6b173').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>param_ct__numerics__scl</th>\n",
       "      <th>param_ct__numerics__ouh</th>\n",
       "      <th>param_clf__tol</th>\n",
       "      <th>param_clf__solver</th>\n",
       "      <th>param_clf__class_weight</th>\n",
       "      <th>param_clf__alpha</th>\n",
       "      <th>param_clf</th>\n",
       "      <th>param_clf__penalty</th>\n",
       "      <th>param_clf__learning_rate</th>\n",
       "      <th>param_clf__l1_ratio</th>\n",
       "      <th>param_clf__eta0</th>\n",
       "      <th>param_clf__early_stopping</th>\n",
       "      <th>param_clf__C</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>MinMaxScaler()</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>newton-cg</td>\n",
       "      <td>balanced</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LogisticRegression(C=1000.0, class_weight='balanced', max_iter=4000,\\n                   random_state=42, solver='newton-cg')</td>\n",
       "      <td>l2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>0.788803</td>\n",
       "      <td>0.031984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>StandardScaler()</td>\n",
       "      <td>None</td>\n",
       "      <td>0.001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.001</td>\n",
       "      <td>SGDClassifier(loss='log_loss', max_iter=4000, random_state=42)</td>\n",
       "      <td>l1</td>\n",
       "      <td>adaptive</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.01</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.788043</td>\n",
       "      <td>0.033164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>MinMaxScaler()</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.01</td>\n",
       "      <td>SGDClassifier(loss='log_loss', max_iter=4000, random_state=42)</td>\n",
       "      <td>l1</td>\n",
       "      <td>adaptive</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.01</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.786998</td>\n",
       "      <td>0.028790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>MinMaxScaler()</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{0: 1, 1: 2}</td>\n",
       "      <td>0.00001</td>\n",
       "      <td>SGDClassifier(loss='log_loss', max_iter=4000, random_state=42)</td>\n",
       "      <td>l1</td>\n",
       "      <td>invscaling</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.01</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.784904</td>\n",
       "      <td>0.030160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>MinMaxScaler()</td>\n",
       "      <td>OutlierHandler(featureindices=[0, 3])</td>\n",
       "      <td>0.001</td>\n",
       "      <td>NaN</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.01</td>\n",
       "      <td>SGDClassifier(loss='log_loss', max_iter=4000, random_state=42)</td>\n",
       "      <td>l2</td>\n",
       "      <td>adaptive</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.01</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.784457</td>\n",
       "      <td>0.030083</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "   param_ct__numerics__scl                param_ct__numerics__ouh  \\\n",
       "73          MinMaxScaler()                                   None   \n",
       "92        StandardScaler()                                   None   \n",
       "51          MinMaxScaler()                                   None   \n",
       "48          MinMaxScaler()                                   None   \n",
       "61          MinMaxScaler()  OutlierHandler(featureindices=[0, 3])   \n",
       "\n",
       "   param_clf__tol param_clf__solver param_clf__class_weight param_clf__alpha  \\\n",
       "73         0.0001         newton-cg                balanced              NaN   \n",
       "92          0.001               NaN                balanced            0.001   \n",
       "51         0.0001               NaN                balanced             0.01   \n",
       "48         0.0001               NaN            {0: 1, 1: 2}          0.00001   \n",
       "61          0.001               NaN                balanced             0.01   \n",
       "\n",
       "                                            param_clf param_clf__penalty  \\\n",
       "73  LogisticRegression(C=1000.0, class_weight='bal...                 l2   \n",
       "92  SGDClassifier(loss='log_loss', max_iter=4000, ...                 l1   \n",
       "51  SGDClassifier(loss='log_loss', max_iter=4000, ...                 l1   \n",
       "48  SGDClassifier(loss='log_loss', max_iter=4000, ...                 l1   \n",
       "61  SGDClassifier(loss='log_loss', max_iter=4000, ...                 l2   \n",
       "\n",
       "   param_clf__learning_rate param_clf__l1_ratio param_clf__eta0  \\\n",
       "73                      NaN                 NaN             NaN   \n",
       "92                 adaptive                 0.5            0.01   \n",
       "51                 adaptive                0.15            0.01   \n",
       "48               invscaling                 0.5            0.01   \n",
       "61                 adaptive                0.15            0.01   \n",
       "\n",
       "   param_clf__early_stopping param_clf__C  mean_test_score  std_test_score  \n",
       "73                       NaN       1000.0         0.788803        0.031984  \n",
       "92                     False          NaN         0.788043        0.033164  \n",
       "51                     False          NaN         0.786998        0.028790  \n",
       "48                     False          NaN         0.784904        0.030160  \n",
       "61                     False          NaN         0.784457        0.030083  "
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ml.gridsearch_to_df(rs1,5)    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=605e84df-e6c3-4f95-86ae-ac3791f0fc5c style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('605e84df-e6c3-4f95-86ae-ac3791f0fc5c').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>MAX of mean_test_score</th>\n",
       "      <th>MIN of mean_fit_time</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>param_clf</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LogisticRegression</th>\n",
       "      <td>0.788803</td>\n",
       "      <td>0.633601</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SGDClassifier</th>\n",
       "      <td>0.788043</td>\n",
       "      <td>0.548365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RidgeClassifier</th>\n",
       "      <td>0.407925</td>\n",
       "      <td>1.017728</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "                    MAX of mean_test_score  MIN of mean_fit_time\n",
       "param_clf                                                       \n",
       "LogisticRegression                0.788803              0.633601\n",
       "SGDClassifier                     0.788043              0.548365\n",
       "RidgeClassifier                   0.407925              1.017728"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ml.compareEstimatorsInGridSearch(rs1,tableorplot='table')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ml.compareEstimatorsInGridSearch(rs1,tableorplot='plot',figsize=(4,4))        "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "4 puanlık bir kaybımız olmuş(notebook her çalışıtğında burdaki değer de değişecektir), o yüzden halvingli versiyon süre açısından faydalı olurken performans açısından faydalı olmadı. Ama şimdi şunu yapabiliriz. rs1'in sonuçlarına bakıp rakamsal parametrelerden en iyi olanların yeni değer uzayını daraltıp şimdi bu sefer halvingliyi kullanabiliriz."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "n_iterations: 3\n",
      "n_required_iterations: 3\n",
      "n_possible_iterations: 3\n",
      "min_resources_: 50\n",
      "max_resources_: 668\n",
      "aggressive_elimination: False\n",
      "factor: 3\n",
      "----------\n",
      "iter: 0\n",
      "n_candidates: 13\n",
      "n_resources: 50\n",
      "Fitting 50 folds for each of 13 candidates, totalling 650 fits\n",
      "----------\n",
      "iter: 1\n",
      "n_candidates: 5\n",
      "n_resources: 150\n",
      "Fitting 50 folds for each of 5 candidates, totalling 250 fits\n",
      "----------\n",
      "iter: 2\n",
      "n_candidates: 2\n",
      "n_resources: 450\n",
      "Fitting 50 folds for each of 2 candidates, totalling 100 fits\n",
      "Wall time: 1min 38s\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-3 {color: black;background-color: white;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-3 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-3 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>HalvingRandomSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                      error_score=&#x27;raise&#x27;,\n",
       "                      estimator=Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                                                 FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                                                     kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                                                (&#x27;ct&#x27;,\n",
       "                                                 ColumnTransformer(n_jobs=-1,\n",
       "                                                                   remainder=&#x27;passthrough&#x27;,\n",
       "                                                                   transformers=[(&#x27;nominals&#x27;,\n",
       "                                                                                  Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                                                   OneHotEncod...\n",
       "                                            &#x27;clf__eta0&#x27;: [0.0005, 0.001, 0.002,\n",
       "                                                          0.005],\n",
       "                                            &#x27;clf__l1_ratio&#x27;: [0.9, 1],\n",
       "                                            &#x27;clf__learning_rate&#x27;: [&#x27;constant&#x27;,\n",
       "                                                                   &#x27;optimal&#x27;,\n",
       "                                                                   &#x27;invscaling&#x27;,\n",
       "                                                                   &#x27;adaptive&#x27;],\n",
       "                                            &#x27;clf__penalty&#x27;: [&#x27;l1&#x27;, &#x27;l2&#x27;,\n",
       "                                                             &#x27;elasticnet&#x27;],\n",
       "                                            &#x27;clf__tol&#x27;: [0.0005, 0.001, 0.002,\n",
       "                                                         0.005],\n",
       "                                            &#x27;ct__numerics__ouh&#x27;: [OutlierHandler(featureindices=[0,\n",
       "                                                                                                 3]),\n",
       "                                                                  None],\n",
       "                                            &#x27;ct__numerics__scl&#x27;: [StandardScaler(),\n",
       "                                                                  MinMaxScaler()]}],\n",
       "                      scoring=&#x27;accuracy&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-29\" type=\"checkbox\" ><label for=\"sk-estimator-id-29\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">HalvingRandomSearchCV</label><div class=\"sk-toggleable__content\"><pre>HalvingRandomSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                      error_score=&#x27;raise&#x27;,\n",
       "                      estimator=Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                                                 FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                                                     kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                                                (&#x27;ct&#x27;,\n",
       "                                                 ColumnTransformer(n_jobs=-1,\n",
       "                                                                   remainder=&#x27;passthrough&#x27;,\n",
       "                                                                   transformers=[(&#x27;nominals&#x27;,\n",
       "                                                                                  Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                                                   OneHotEncod...\n",
       "                                            &#x27;clf__eta0&#x27;: [0.0005, 0.001, 0.002,\n",
       "                                                          0.005],\n",
       "                                            &#x27;clf__l1_ratio&#x27;: [0.9, 1],\n",
       "                                            &#x27;clf__learning_rate&#x27;: [&#x27;constant&#x27;,\n",
       "                                                                   &#x27;optimal&#x27;,\n",
       "                                                                   &#x27;invscaling&#x27;,\n",
       "                                                                   &#x27;adaptive&#x27;],\n",
       "                                            &#x27;clf__penalty&#x27;: [&#x27;l1&#x27;, &#x27;l2&#x27;,\n",
       "                                                             &#x27;elasticnet&#x27;],\n",
       "                                            &#x27;clf__tol&#x27;: [0.0005, 0.001, 0.002,\n",
       "                                                         0.005],\n",
       "                                            &#x27;ct__numerics__ouh&#x27;: [OutlierHandler(featureindices=[0,\n",
       "                                                                                                 3]),\n",
       "                                                                  None],\n",
       "                                            &#x27;ct__numerics__scl&#x27;: [StandardScaler(),\n",
       "                                                                  MinMaxScaler()]}],\n",
       "                      scoring=&#x27;accuracy&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-30\" type=\"checkbox\" ><label for=\"sk-estimator-id-30\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                 FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                     kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                (&#x27;ct&#x27;,\n",
       "                 ColumnTransformer(n_jobs=-1, remainder=&#x27;passthrough&#x27;,\n",
       "                                   transformers=[(&#x27;nominals&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                   OneHotEncoder(drop=&#x27;first&#x27;,\n",
       "                                                                                 handle_unknown=&#x27;ignore&#x27;))]),\n",
       "                                                  [1, 6, 7]),\n",
       "                                                 (&#x27;numerics&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;imp&#x27;,\n",
       "                                                                   SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                                                                  (&#x27;ouh&#x27;,\n",
       "                                                                   DummyTransformer()),\n",
       "                                                                  (&#x27;scl&#x27;,\n",
       "                                                                   DummyTransformer())]),\n",
       "                                                  [2, 3, 4, 5])])),\n",
       "                (&#x27;fs&#x27;,\n",
       "                 SelectKBest(score_func=&lt;function mutual_info_classif at 0x000001BC26E09B80&gt;)),\n",
       "                (&#x27;clf&#x27;, DummyEstimator())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-31\" type=\"checkbox\" ><label for=\"sk-estimator-id-31\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                    kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})</pre></div></div></div><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-32\" type=\"checkbox\" ><label for=\"sk-estimator-id-32\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">ct: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(n_jobs=-1, remainder=&#x27;passthrough&#x27;,\n",
       "                  transformers=[(&#x27;nominals&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                  OneHotEncoder(drop=&#x27;first&#x27;,\n",
       "                                                                handle_unknown=&#x27;ignore&#x27;))]),\n",
       "                                 [1, 6, 7]),\n",
       "                                (&#x27;numerics&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;imp&#x27;,\n",
       "                                                  SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                                                 (&#x27;ouh&#x27;, DummyTransformer()),\n",
       "                                                 (&#x27;scl&#x27;, DummyTransformer())]),\n",
       "                                 [2, 3, 4, 5])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-33\" type=\"checkbox\" ><label for=\"sk-estimator-id-33\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">nominals</label><div class=\"sk-toggleable__content\"><pre>[1, 6, 7]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-34\" type=\"checkbox\" ><label for=\"sk-estimator-id-34\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder(drop=&#x27;first&#x27;, handle_unknown=&#x27;ignore&#x27;)</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-35\" type=\"checkbox\" ><label for=\"sk-estimator-id-35\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">numerics</label><div class=\"sk-toggleable__content\"><pre>[2, 3, 4, 5]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-36\" type=\"checkbox\" ><label for=\"sk-estimator-id-36\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;median&#x27;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-37\" type=\"checkbox\" ><label for=\"sk-estimator-id-37\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyTransformer</label><div class=\"sk-toggleable__content\"><pre>DummyTransformer()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-38\" type=\"checkbox\" ><label for=\"sk-estimator-id-38\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyTransformer</label><div class=\"sk-toggleable__content\"><pre>DummyTransformer()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-39\" type=\"checkbox\" ><label for=\"sk-estimator-id-39\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">remainder</label><div class=\"sk-toggleable__content\"><pre></pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-40\" type=\"checkbox\" ><label for=\"sk-estimator-id-40\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">passthrough</label><div class=\"sk-toggleable__content\"><pre>passthrough</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-41\" type=\"checkbox\" ><label for=\"sk-estimator-id-41\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SelectKBest</label><div class=\"sk-toggleable__content\"><pre>SelectKBest(score_func=&lt;function mutual_info_classif at 0x000001BC26E09B80&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-42\" type=\"checkbox\" ><label for=\"sk-estimator-id-42\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyEstimator</label><div class=\"sk-toggleable__content\"><pre>DummyEstimator()</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "HalvingRandomSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                      error_score='raise',\n",
       "                      estimator=Pipeline(steps=[('log',\n",
       "                                                 FunctionTransformer(func=<function logTransformer at 0x000001BC2B98CE50>,\n",
       "                                                                     kw_args={'col_name': 'Fare'})),\n",
       "                                                ('ct',\n",
       "                                                 ColumnTransformer(n_jobs=-1,\n",
       "                                                                   remainder='passthrough',\n",
       "                                                                   transformers=[('nominals',\n",
       "                                                                                  Pipeline(steps=[('ohe',\n",
       "                                                                                                   OneHotEncod...\n",
       "                                            'clf__eta0': [0.0005, 0.001, 0.002,\n",
       "                                                          0.005],\n",
       "                                            'clf__l1_ratio': [0.9, 1],\n",
       "                                            'clf__learning_rate': ['constant',\n",
       "                                                                   'optimal',\n",
       "                                                                   'invscaling',\n",
       "                                                                   'adaptive'],\n",
       "                                            'clf__penalty': ['l1', 'l2',\n",
       "                                                             'elasticnet'],\n",
       "                                            'clf__tol': [0.0005, 0.001, 0.002,\n",
       "                                                         0.005],\n",
       "                                            'ct__numerics__ouh': [OutlierHandler(featureindices=[0,\n",
       "                                                                                                 3]),\n",
       "                                                                  None],\n",
       "                                            'ct__numerics__scl': [StandardScaler(),\n",
       "                                                                  MinMaxScaler()]}],\n",
       "                      scoring='accuracy', verbose=1)"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "params= [\n",
    "    {\n",
    "           'clf'               : [SGDClassifier(loss='log_loss',max_iter=mi,random_state=42)],\n",
    "           'clf__alpha'        : [0.005, 0.01, 0.02, 0.05], \n",
    "           'clf__tol'          : [0.0005, 0.001,0.002,0.005],                             \n",
    "           'clf__penalty'      : ['l1','l2','elasticnet'],    \n",
    "           'clf__l1_ratio'     : [0.9, 1],   \n",
    "           'clf__learning_rate': ['constant','optimal','invscaling','adaptive'],\n",
    "           'clf__eta0'          : [0.0005, 0.001, 0.002, 0.005],    \n",
    "           'clf__early_stopping':[True,False], \n",
    "           'clf__class_weight': ['balanced'],                \n",
    "           'ct__numerics__ouh': [OutlierHandler(featureindices=[0,3]),None], #Age ve Fare için, her ne kadar Fare'de LogTrans yapmış olsak da\n",
    "           'ct__numerics__scl' : [StandardScaler(),MinMaxScaler()]\n",
    "          }   \n",
    "] \n",
    "\n",
    "hrs2 = HalvingRandomSearchCV(estimator = pipe, param_distributions = params, cv = mycv, n_jobs=-1, verbose = 1, \n",
    "                           scoring = 'accuracy',error_score='raise',min_resources=min_res,factor=fact) \n",
    "\n",
    "hrs2.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7899675405742821"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hrs2.best_score_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Çok da işe yaramadı, ilk buluduğmuz sonuçlar daha iyiydi, belki birkaç kez çalıştırmak gerekebilir. Bununla birlikte \"titanic logistic regression\" diye bir google araması yaparsanız zaten çoğu kişin %80 civarı bir accuracy bulduğunu görürsünüz. Başka neler yapılarak skor iyileştirilebilir:\n",
    "\n",
    "- kişilerin ismindeki Mr., Mrs gibi ünvanlar çıkartılabilir\n",
    "- yukarda söylediğimiz ama yapmadığımız, age için minimum yaş olarak 2 belirlenebilir, hatta age alanı belki discretize edilebilir\n",
    "- fare featuru da discretize edilebilir\n",
    "- age alanının simpleimputer yapmak yerine, iterativeimputer yapılabilir. yukarıdaki pipelineımız bunu destekler nitelikte, orada aynı işlem için istediğimiz kadar transformer kullanabiliyoruz(scalerlarda yaptığmız gibi)\n",
    "- ve şuan aklıma gelmeyen, sizin aklınıza gelebilecek diğer yöntemler"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model değerlendirme(evaluation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ml.plot_learning_curve(rs1.best_estimator_,\"Learnig curve\",X_train,y_train,cv=mycv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUEAAAEQCAYAAAAu3MZjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAgcklEQVR4nO3deZwdVZn/8c836ZCQFZImEVkSQCEaNCgZAZVBDQo4joNEBUUFFIPww1EZUVTQyDKKKG5RMQqiAgrMAAI6IBlZXAANCsFIzBAkEpYkTUJ2kl6e3x9VHS7XTnf1vdV36fq+X696cetUddXT3eTpc+qcOkcRgZlZUQ2pdwBmZvXkJGhmheYkaGaF5iRoZoXmJGhmheYkaGaF5iRoZoXmJGhmDUXS6ZIWSNoi6fKyYzMlLZa0SdLtkiaXHBsu6TJJ6yQ9JemMLPdryTn+3LWOHxpT9hhW7zCsH/76t9Z6h2D98Oyza2jfulHVXOOI14+Kp1d3Zjr3voVbbo2II3s55QngfOAIYMfuQkmtwHXAycBNwHnA1cDB6SlzgBcDk4EXALdL+ktE3NJbPA2fBKfsMYzf37pHvcOwfnjD+z5Q7xCsH+67Z27V12hb3cm9t+6e6dxhuy7t9a9kRFwHIGkGUHrRY4BFEXFtenwO0CZpakQsBk4AToyINcAaSd8DTgSaOwmaWTMIOqNroG8yDXhg2x0jNkpaCkyTtALYtfR4+vnovi7qJGhmVQugi8zzELRKWlCyPy8i5mX4utHAqrKytcCY9Fj3fvmxXjkJmlkuushcE2yLiBkV3GIDMLasbCywPj3Wvf9s2bFeuXfYzKoWBO3RlWmrwiJgeveOpFHAPiTPCdcAT5YeTz8v6uuiToJmVrUAOolMW18ktUgaAQwFhkoaIakFuB7YX9Ks9PhngYVppwjAj4CzJe0saSrwQeDyvu7nJGhmuegiMm0ZnA1sBs4C3pN+PjsiVgGzgAuANcBBwHElX/c5YCmwDLgTuKiv4THgZ4JmloMAOnOaoDki5pCM+evp2Hxg6naObQHen26ZOQmaWS4GfIDMAHESNLOqRcbnfY3ISdDMqhYB7c2ZA50EzSwPopOqXj+uGydBM6taAF2uCZpZkbkmaGaFlQyWdhI0s4IKoD2a890LJ0Ezq1ogOpv0BTQnQTPLRVe4OWxmBeVngmZWcKLTzwTNrKiSmaWdBM2soCLE1hha7zAq4iRoZrno8jNBMyuqpGPEzWEzK6zm7RhpzqjNrKF0d4xk2bKQ9BJJv5K0VtLDkt5WcmympMWSNkm6XdLkamJ3EjSzXHSGMm19SRdV+hlwMzAemA1cIWlfSa3AdcA56bEFwNXVxO3msJlVLRDtkVs6mQq8EPhqRATwK0m/Bd4LPEayxOa1AJLmAG2SppasOtcvToJmVrUadIwI2J9kQfUHtt03YqOkpcA0oKIk6OawmVUtyNYUTpvDrZIWlGyzyy73V2AlcKakYZLeBBwGjARGA2vLzl8LjKk0dtcEzSwX/XhjpC0iZmzvYES0Szoa+CbwSZLnftcAW4ANJLXBUmOB9f2Nt5uToJlVLYJch8hExEKS2h8Akn4H/JCk5X1CSfkoYB9gUaX3chI0s6olHSP5vTYn6eXAEpJHdqcBuwKXk9T6LpI0C/g58FlgYaWdIuBngmaWk06GZNoyei/wJMmzwZnAGyNiS0SsAmYBFwBrgIOA46qJ2zVBM6taoFwnVY2IM4Ezt3NsPskwmlw4CZpZLvzusJkVVrLusJOgmRWWPL2+mRVXsuSmJ1U1s4KKkJvDZlZszTqfoJOgmVUtmU/QzwTNrLCad2ZpJ0Ezq1oyRMY1QTMrqLzfHa4lJ0Ezy4UXXzezwkqm0nJz2MwKzM8Ezaywkllk3By2El/52B7cO38sO7V2MO/2vwKwbs1Q/vNDU1ixfAcm7b6Vz3z3Ucbs1Mnf/284F5+xJw8/uCMnfPJJ3nHqqjpHX2zDhnXw9U//gmHDOhk6JLjzD1P44fWvBIL3v/0+DnvVo3R1iRv/dyrX3zat3uE2hOS1OSfBPkkaD1wKvAloAz4VEVfVMoZaedOxq3nrSW1c9JE9t5VdM3cir3jteo798Equ/uZErp47kZPPfpKxO3dy6nnL+d0t4+oYsXVrbx/KGV88ime3DGPo0C6+cfbN/H7h7kx+4TNMHL+REz85iwix05jN9Q61gTRvTbDWUX8L2ApMAo4HviNpUP4pfdnBGxmzc+fzyu6+dRyHv3M1AIe/czV3p0lvp9YO9jtgMy2ulzcI8eyWYQC0DO2iZWgQAW99w2J+dMMriPTZ1zPrd6xnkA2nC2XaspA0RdIvJK2R9JSkuemi7Eg6QNJ9kjal/z2gmrhr9s8uXRBlFrB/RGwAfiPpRpJptM+qVRz1tKZtGBMmdQAwfmIHa9qG1Tki254h6uKSc29kt0nruGH+S1j8yER2nbSe1x/0CK+dsYxn1o1g7hUH8/gK195hQHqHv00ytf6uwE7AbcBpki4BfgZ8LT3nFOBnkl4cEVsruVEta4L7Ah0RsaSk7AGSRZMLRwIp6h2GbUdXDGH2OUfzzo8ey9S9VzFltzXs0NLJ1vahnPq5f+MXd+7HmSf/pt5hNpSuGJJpy2gv4JqIeDYingJuIckVryOpvH0tXXPkGyQLs7+h0rhrmQRHA+vKynpcNFnS7O6FmVc93Vl+uGnt3NrO0yuSyvfTK1rYaUJHnSOyvmzcNJz7H9qVV718OatWj+LXC6YA8OsFk9l7j9X1Da6BdK8xkmXL6GvAcZJGStoNOIrnEuHCiCitQSykispULZNg5kWTI2JeRMyIiBm7TGjOV3F6cvCb1jH/mvEAzL9mPIccsbbOEVlPxo3ZzKiRWwDYYVgHB+7/BH9/chy//eOeHPDSJwGYPvUplj/lpnC3ADpiSKYNaO2u5KTb7B4ueRdJYlsHLCdZgP0GkspU+T+cHitTWdXyUfwSoCVtu/9fWjadKhZNbmRfOHUyC+8ezdrVLRx/4Et57388xbGnr+CCD03hlp9OYOJuyRAZgNUrW/jwUfuyaf1QNARu+P4uzLtjMaPGdNX3myioCTtt5pOz72KIgiFDgjvu3Yt77t+TB5dM4jMfupO3H7GIzVta+PKlr6l3qA2lH03dtoiYsb2DkoaQ1PrmAa8mSXyXAReSLMOZqTKVlZ5fqxxYkn5K8kfjZOAA4BfAqyNiu4lwxvQR8ftb96hNgJaLN7zvA/UOwfrhvnvmsn7d8qp6NcZPnRgzL5uV6dz/es0l9/WRBFuBVcBOEbE2LTsaOB84gyQh7tHdJJa0DDglIm6pJPZaD5E5DdiRpNfnJ8CpvSVAM2sO3ZOq5jFEJiLagL8Bp0pqkbQTcALJs787gE7g3yUNl3R6+mW/qjT2mibBiFgdEUdHxKiI2HOwDpQ2K6KcO0aOAY4kqRE+DLQDH0uHwRwNvA94Bng/cHSlw2PAr82ZWQ7ynlQ1Iu4nGQ7T07E/AQfmdS8nQTOrWiA6uprztTknQTPLhRdaMrPiCs8naGYF5oWWzKzwnATNrLAC0emOETMrMneMmFlhhTtGzKzowknQzIqrX6/ENRQnQTPLhWuCZlZYEdDZ5SRoZgXm3mEzK6zAzWEzK7RB2DEi6ZisF4mI6/IJx8yaVQ1X6shVbzXB/8p4jQAGz5JwZlaRZm0Ob/dlv4gYknFzAjQruKR3eEimrS+SNpRtnZK+WXJ8pqTFkjZJul3S5Gpib843ns2s4URk2/q+Tozu3oAXAJuBa2HbSnTXAecA40nWI766mrgzJ0FJR0n6uaSHJO2Rlp0saWY1AZjZ4BChTFs/zSJZnfLX6f4xwKKIuDYingXmANMlTa007kxJUNLxwDUkC6hPAYalh4YCn6j05mY2OATZEmAFSfAE4EfdawwD04AHtt03YiOwNC2vSNaa4CeAD0bEx4COkvJ7SBZRN7OCi4wb0CppQck2u6frpc/6DgN+WFI8GlhbdupaYEylcWcdJ/hi4O4eyjcAYyu9uZkNEgGR/bW5toiYkeG89wK/iYi/lZT1lHPGAuuz3rxc1prgE8C+PZT/M0lV1MwKbgCaw+/j+bVAgEXA9O4dSaOAfdLyimRNgvOAb0h6Tbq/h6QTgC8B36n05mY2eOTVOwwg6dXAbqS9wiWuB/aXNEvSCOCzwMKIWFxp3JmawxHxJUnjgNuAEcDtwBbgyxHxrUpvbmaDwwC8O3wCcF1EPK+ZGxGrJM0C5gJXAPcCx1Vzo8zvDkfEZyRdALyUpAb5l4jYUM3NzWyQCCDHJBgRp/RybD5Q8ZCYcv2dQCGAZ9PPnXkFYWbNr1nfHc46TnC4pK8Bq0nG6CwEVkv6etouN7NCE9GVbWs0WWuC3wHeBJzMc0NlDgG+QDI+5/35h2ZmTaVJa4JZk+A7gGMi4raSskckrQT+GydBs2KL5p1FJmsS3Ag83kP54yQvN5tZ0TVpTTDrOMFvAp+TtGN3Qfr5nPSYmRWeMm6NpbeZpW8sK3od8Likhen+y9KvHzUwoZlZU+mqdwCV6a05/HTZ/n+X7f8NMzPIfZxgLW03CUbESbUMxMyaW7OOE/Rqc2aWj8GeBCWdBLwL2BPYofRYROydc1xm1myatDmc9Y2RM4GvAPeRzCx9A/Bnkjn+Lxug2MysiSiybY0m6xCZDwKzI+JTQDswNyLeSpIYq1rpycwGgRB0ZdwaTNYkuDvw+/TzZp6b2fUnJAuhmFnR9WN+/UaSNQk+BbSmn5eRvDcM8CIa8tsys5ob5EnwV8Bb08+XAhdLup1kvc/rBiIwM2sygzwJzgbOB4iIS4ATgQeBzwCnDUhkZtY8ugdLZ9kyknRcus75RklLJR2als+UtFjSJkm3p6vSVSzr9PpdlLwUExFXU+Wq72Y2uOTZ8yvpjcCFwLEk/RG7puWtJK3Pk4GbgPNIctHBld6rt3eHX5n1IhHxx0oDMLNBIt+m7ueBcyPinnT/cYB0jeJFEXFtuj8HaJM0tdLFlnqrCS4g+bb6qr8GMLSSm5vZ4JFXTVDSUGAGcKOkh0kWd7sBOBOYRjK7PQARsVHS0rQ89yS4VyUXzNuShSM54oUH1DsM64dlP27S6UQKauvDOWWv7M/7WiUtKNmfFxHzSvYnAcOAtwOHkoxN/hlwNjAaWFV2vbUkM9xXpLcJFJZVelEzK5j+9fy2RcSMXo53T9T8zYh4EkDSxSRJ8C6eG6fcbSywngpl7R02M+tdTkNkImINsLzs7O7Pi4Dp3YWSRgH7pOUVcRI0s1yoK9uW0Q+AD0uaKGln4GPAzcD1wP6SZqUrXX4WWFhppwg4CZpZXvIdLH0e8AdgCfAQ8CfggohYRfKq7gXAGuAg4LhqwvZ8gmZWtbxniImIdpIXMf7hZYyImA9Mzete/aoJSmqVdJCk4XkFYGaDRM5vjNRK1vkEx0i6BlgJ/A7YLS2/JB2saGZFN8jfHb6QJPG9kuevM3wz8La8gzKz5tOsk6pmfSb4VuBtEXG/9Lxv4yHAU+ubFV30q+e3oWRNgjvzj0twQjJKuzO/cMysaTVgLS+LrM3hP/DcfILw3Ld7CskzQjMruiZ9Jpi1Jvhp4FZJ09KvOSP9/CrgnwcqODNrHo34vC+LTDXBiPgd8GqSpTaXAjOBJ4BDPI2WmTWzzIOlI+JB4IQBjMXMmlmT1gQzJUFJ43s7HhGr8wnHzJpSAXqH2+g9z3tSVbOiG8w1QeD1ZfvDgFcAp5LM8WVmBSaat2Mk60JLd/ZQPF/SIyQLnlyVa1Rm1nwGcxLsxf14iIyZNegrcVlUnAQljQY+CjyWWzRm1rwGc8eIpPU8v7IrYCSwETh+AOIysyYz2GuCp5ftd5Gs+HRvuh6AmRXdYE2CklqAUcANEfHEwIdkZk0n5/eCJd0BHAx0pEWPR8R+6bF3A18AWoHbgPdXM1a5z9fmIqIDuIhkWIyZWY8GYD7B0yNidLp1J8BpwHeB95KsT7wJ+HY1cWdtDt8DHAh4LWIz61ltmsPHAzdFxF0Aks4BHpI0JiIqWns4axL8HvBlSXsC95F0iGzjSRTMbABem/uCpC8CfwU+ExF3ANMomb4vIpZK2grsS5Kb+q3XJCjpMpJhMN2DoS/u4bTAr82ZFVv/ngm2SlpQsj8vIuaVnfNJ4C/AVpIlNW+SdAAwGlhbdu5akgmeK9JXTfAE4Cxgr0pvYGaDn9Ito7aImNHbCRFxb8nuDyW9C3gzsAEYW3b6WKCipjD0nQSVBuRngWbWu4F9Jhgk+WgRML27UNLewHCSRdorkuWZYJOO/jGzWsprsLSknYCDgDtJhsgcS/J67kdIRqncLelQ4I/AucB1lXaKQLYk+JTUe0U3IvxM0Kzo8qsuDQPOB6aSLOS2GDg6IpYASPoQcCUwAZgPnFTNzbIkwdnAM9XcxMwGuRwnVY2IVcA/9XL8KnKcuSpLErwpIlbmdUMzG6Sa9MFZX0mwSb8tM6u1wTqBQj96vc2s0AZjEoyIrIuzm1nBDdaaoJlZ34LBPamqmVlvBv1CS2ZmfXISNLMiUzRnFnQSNLPq5TyzdC05CZpZLvxM0MwKbQAmVa0JJ0Ezy4drgmZWWP1fRKlhOAmaWT6cBM2sqDxY2swKT13NmQU9QYKZVS/6sWUk6cWSnpV0RUnZuyUtk7RR0g2SxlcbumuCNXDGxX/noMPX80xbC6e8YT8A9n7pZj78xeXsOKqLFct34ML/tyebNniVgkYxZGMHEy99jB2WbwbBypMnM+7Wlezw5Jbk+KZOukYO5bELptY50sYxAENkvgX8Ydv1pWnAd4F/IVlfZB7wbZIlOStW0yQo6XTgROBlwE8i4sRa3r9efnn1eG78QStnfv2xbWUf/fJjfO/cF/LgPaN503FP8/ZTV/Kji3atY5RWqvWKx9n08jE89e97QUcXQ7Z0seL051aenXDVcrp29B+t58mxNSzpOJJlPX4HvCgtPp5kpvu70nPOAR6SNKaahZZq3Rx+gmQBlctqfN+6+vO9o1m/5vl/b3bfewsP3jMKgD/dNYbX/kv5etJWL0M2dbLj4g2sO2xCUtAyhK5RJb+/CEbf+wwbDtm5PgE2KEW2rc/rSGNJVpE7o+zQNOCB7p2IWEqyOPu+1cRd05pgRFwHIGkGsHst791oli0ZwSFHruPuW8Zx6FvWsssL2+sdkqVaVm2hc2wLE+f9neGPbebZKSNpe89uxIik5jfirxvpHNdC+wtG1DnSBhJA9gkUWiUtKNmfFxHzSvbPAy6NiOVlK12OBsprC2uBMf2M9nkasmNE0mxJCyQtaGdLvcMZEBefsQf/ekIbc29Zwo6jO+nY6pUMGoU6Yfijm1g7s5XHzp9KDB/Czjev2HZ8zN1r2HCwa4Hl1JVtA9oiYkbJti0BSjoAOBz4ag+32ACMLSsbC1TcFIYG7RhJfyjzAMZqfHP2u/fhsYdH8Ol37QPAbntv4aCZ6+ockXXrGD+MjvE7sOVFyeOKDa/aiZ1vSpNgZzBqwTM8dt5+dYyw8eQ4TvB1wBTg72ktcDQwVNJLgVuA6dvuKe0NDAeWVHPDhkyCRTBuQjtrnx6GFLz7Iyu4+ccT6h2SpTp3GkbH+GEMe/JZ2ncdwchF69m6W9L0HbloPe27jqBz/A51jrLBRPSnOdybecBPS/Y/TpIUTwUmAndLOpSkd/hc4LpqOkXASbAmzvr2Ml5+yAbGje/gigV/4cdfmcSOI7v41xPbAPjt/4zjlz+teriT5WjV+3Zn0nceRR1B+y7DWTl7TwBG372G9e4Q6VEeNcGI2ARs2nZNaQPwbLog+ypJHwKuBCYA84GTqr1nrYfItKT3HEpSxR0BdERERy3jqLUvnja5x/IbLt2lxpFYVlsnj2T5uf84BnDlKT3/Lo0BeXc4IuaU7V8FXJXnPWrdMXI2sBk4C3hP+vnsGsdgZgMgryEytVbrITJzgDm1vKeZ1UAAnQ2Y4TLwM0Ezy0Uj1vKycBI0s3x4tTkzKzLXBM2suLzkppkVmQC5Y8TMikx+JmhmheXmsJkVW27vDteck6CZ5cK9w2ZWbK4JmllhhXuHzazomjMHOgmaWT48RMbMis1J0MwKK4D8F1+viYZcbc7MmosIFNm2TNeTrpD0pKR1kpZIOrnk2ExJiyVtknS7pKqm+3YSNLN8dHVl27L5AjAlIsYCbwXOl3SgpFbgOuAcYDywALi6mrDdHDaz6uXcHI6IRWVXD2Af4EBgUURcCyBpDtAmaWpELK7kXq4Jmlku8mwOA0j6tqRNwGLgSeAXwDTgge5zImIjsDQtr4iToJnlo3vt4b42aJW0oGSb3fPl4jRgDHAoSRN4C8li7GvLTl2bnlcRN4fNLAf9mkChLSJmZLpqRCfwG0nvIVmAfQMwtuy0sUDFC7C7Jmhm1etebS7LVpkWkmeCi4Dp3YWSRpWUV8RJ0MxykdczQUkTJR0nabSkoZKOAN4F/C9wPbC/pFmSRgCfBRZW2ikCbg6bWV7ye2MkSJq+l5BU1JYBH42IGwEkzQLmAlcA9wLHVXMzJ0Ezq14AXfkkwYhYBRzWy/H5wNRcboaToJnlwjNLm1nROQmaWWEF0NmcMyg4CZpZDgLCSdDMiszNYTMrrBx7h2vNSdDM8uGaoJkVmpOgmRVWBHR21juKijgJmlk+XBM0s0JzEjSz4gr3DptZgQWEB0ubWaH5tTkzK6yI/iyn2VCcBM0sH03aMeLp9c0sF9HVlWnri6Thki6VtEzSekn3Szqq5PhMSYslbZJ0u6TJ1cTtJGhmOci43Ga22mIL8BjJ7NLjgLOBayRNkdRKsvzmOcB4YAFwdTWRuzlsZtXLd3r9jcCckqKbJf0NOBCYACyKiGsBJM0B2iRNrXSxJdcEzaxqAURnZ6atvyRNAvYlWVZzGvDAtvsmCXNpWl4R1wTNrHrRr0lVWyUtKNmfFxHzejpR0jDgSuCHEbFY0mhgVdlpa4Ex/Q25m5OgmeUisjeH2yJiRl8nSRoC/BjYCpyeFm8AxpadOhZYn/Xm5dwcNrN8RFe2LQNJAi4FJgGzIqI9PbQImF5y3ihgn7S8IooGH9sjaRXJ4suDTSvQVu8grF8G6+9sckTsUs0FJN1C8vPJoi0ijuzjepcABwCHR8SGkvJdgIeB9wM/Bz4PHBYRB1cSNzRBEhysJC3I0iSwxuHfWW2k4/4eBbYAHSWHTomIKyUdDswFJgP3AidGxKOV3s/PBM2soUTEMkC9HJ8PTM3rfn4maGaF5iRYPz0OCbCG5t/ZIORngmZWaK4JmlmhOQmaWaE5CdaYpPGSrpe0MZ0q6N31jsm2T9LpkhZI2iLp8nrHY/nzEJna+xbJa0CTSAaD/lzSAxFR8Yh3G1BPAOcDRwA71jkWGwDuGKmh9BWfNcD+EbEkLfsx8HhEnFXX4KxXks4Hdo+IE+sdi+XLzeHa2hfo6E6AqQeoYhogM6uOk2BtjQbWlZVVNQ2QmVXHSbC2cp8GyMyq4yRYW0uAFkkvLimbThXTAJlZdZwEayidCvw64FxJoyS9Bvg3kokjrQFJapE0AhgKDJU0QpJHVQwiToK1dxrJUIuVwE+AUz08pqGdDWwGzgLek34+u64RWa48RMbMCs01QTMrNCdBMys0J0EzKzQnQTMrNCdBMys0J0EzKzQnwYKR9HZJUbJ/oqQNvX3NAMZyc29z9El6naSQlHU9WyTdIWlulXFNSe/r5TULwEmwAUi6PP1HF5LaJT0i6cvp1FsD7Wpg76wnS3pU0scHMB6zmvLrP41jPvBeYBhwKPB9YBRwavmJ6WtbnZHDSPeI2EzyFoRZIbkm2Di2RMRTEfFYRFwFXAkcDSBpjqQ/p03XpcAWYJSkcZLmSVopab2kO8ubcJLel07jv0nSzSQzWpce/4fmsKQ3S7pX0mZJT0u6KX1n9g5gMnBRd8215Gtend5/k6THJX1H0tiS4yPTGu8GSSskfbq/PyBJEyT9RNLyNLZFkk7q4dQWSV+XtCbdLpI0pOQ6O0i6ML3OJkl/kHREf+OxwcFJsHFtJqkVdtsLeDfwDpKZZ7YAPwd2A94CvAK4C/iVpF0BJB0EXE6yXu4BwE3Aub3dVNKRwI3AbcCBwOuBO0n+XzkGWJ5eY9d0Q9LLgF+mXzc9Pe8A4LKSS38ZeCMwC5iZxvvPmX8aiRHAH9PvdxrwdeC7kmaWnXd8Gu8hwCnAbOCjJcd/ABxG8vPcH/ghcJOk6f2MxwaDiPBW540kUd1csv8qoA24Ot2fA7QDk0rOeQPJ/IQ7ll3rfuAT6eergNvKjn8/+bVv2z8R2FCy/1vgp73E+ijw8bKyHwGXlpUdAAQwkWQy2S3A8SXHRwPPAJf3cq/Xpddo7eWcnwLfL9m/g2TKMpWUnQ0sTz/vA3QBe5Zd5wbg2+nnKel9Z9T7/w1vA7/5mWDjODJtlraQ1AB/Bny45PjyiFhRsn8gMBJYJan0OiNI/qEDvISk9lfqbuADvcTxCpKk3B8HAi+SdGxJWXdQ+wCbgB3SewMQERskPdifm0gaSjKby7EkNeDh6XXvKDv1nkizWepu4Ly0ef7KNLa/lP3chgO/6k88Njg4CTaOu0iabe3AExHRXnZ8Y9n+EGAFSSdKufIp/AfaEJIa5ld7OPY4ydoqefg48B/AR4AHSWrC/0lS28xqCEkt759Iftal3EFUQE6CjWNTRDzcj/P/SNLJ0RURj2znnIeAg8vKyvfL/Ynkmd33tnN8K8kEo+WxTNte/GlnTnt670fSslEkz+OW9hFPqdcCN0XEj9NriCTBPlN23kGSVFIbPJjkD8s6SX8iqQm+ICJu78e9bZByx0jzmk/y/O5nko6StJekQyR9XlJ37fAbwOGSPiXpxZI+CLytj+teALxD0vmSXippmqSPSRqZHn8UOFTSbiWDmC8EXiXpEkmvkPQiSW+R9F1Imr7ApcCFkt4oaRpJp0l5Mu3LEmCmpNdKmgrMJekwKvdC4GuS9pP0duBM0lpqJCv9XQlcrmTg+N6SZkj6uKRj+hmPDQJOgk0qreW8meQ51veAvwLXAPuRLBhORNxD8vzvVGAhSa/tnD6u+wuSRHkUSa3wTpIe4q70lM8Ce5DU4FalX7OQpKd3Snr+A8AXSJrr3T4O3A5cn/73zySPAPrjfOD3wP+kX7uRJKGVu5Ikwd5L8rO5lOc31U8i6SH+ErAYuDmNf1k/47FBwDNLm1mhuSZoZoXmJGhmheYkaGaF5iRoZoXmJGhmheYkaGaF5iRoZoXmJGhmheYkaGaF9v8BPcPSw4k9VIAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
    "y_pred=rs1.predict(X_test)\n",
    "cm = confusion_matrix(y_test,y_pred)\n",
    "ConfusionMatrixDisplay(cm,display_labels=rs1.classes_).plot();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.84      0.74      0.79       137\n",
      "           1       0.65      0.78      0.71        86\n",
      "\n",
      "    accuracy                           0.75       223\n",
      "   macro avg       0.75      0.76      0.75       223\n",
      "weighted avg       0.77      0.75      0.76       223\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "print(classification_report(y_test,y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#if not imbalanced\n",
    "ml.plotROC(y_test, X_test, rs1, pos_label=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Eğer olur da best_estimator'ün predict_proba metodu yoksa(mesela RidgeClasffier), o zaman bir başkasına bakabiliriz."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rs1.set_params(**ml.getAnotherEstimatorFromGridSearch(rs1,\"SGDClassifier\")[0]).fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ml.plot_precision_recall_curve(y_test,X_test,rs1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\volka\\appdata\\local\\programs\\python\\python38\\lib\\site-packages\\sklearn\\utils\\deprecation.py:87: FutureWarning: Function plot_precision_recall_curve is deprecated; Function `plot_precision_recall_curve` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: PrecisionRecallDisplay.from_predictions or PrecisionRecallDisplay.from_estimator.\n",
      "  warnings.warn(msg, category=FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.precision_recall_curve.PrecisionRecallDisplay at 0x1bc2d5bb6a0>"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import plot_precision_recall_curve, PrecisionRecallDisplay\n",
    "plot_precision_recall_curve(rs1.best_estimator_,X_test,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.precision_recall_curve.PrecisionRecallDisplay at 0x1bc2d67d940>"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pred_prob = rs1.best_estimator_.predict_proba(X_test) \n",
    "precision, recall, thresholds = precision_recall_curve(y_test, pred_prob[:,1])\n",
    "PrecisionRecallDisplay(precision, recall,pos_label=1).plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=c963beae-9b77-4cfc-b747-ca11e184db14 style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('c963beae-9b77-4cfc-b747-ca11e184db14').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Decile</th>\n",
       "      <th>Number of Cases</th>\n",
       "      <th>Number of Responses</th>\n",
       "      <th>Cumulative Responses</th>\n",
       "      <th>% of Events</th>\n",
       "      <th>Gain</th>\n",
       "      <th>lift</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>22</td>\n",
       "      <td>22</td>\n",
       "      <td>25.58</td>\n",
       "      <td>25.58</td>\n",
       "      <td>2.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>22</td>\n",
       "      <td>16</td>\n",
       "      <td>38</td>\n",
       "      <td>18.60</td>\n",
       "      <td>44.18</td>\n",
       "      <td>2.21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>22</td>\n",
       "      <td>13</td>\n",
       "      <td>51</td>\n",
       "      <td>15.12</td>\n",
       "      <td>59.30</td>\n",
       "      <td>1.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>22</td>\n",
       "      <td>13</td>\n",
       "      <td>64</td>\n",
       "      <td>15.12</td>\n",
       "      <td>74.42</td>\n",
       "      <td>1.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>22</td>\n",
       "      <td>6</td>\n",
       "      <td>70</td>\n",
       "      <td>6.98</td>\n",
       "      <td>81.40</td>\n",
       "      <td>1.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>23</td>\n",
       "      <td>7</td>\n",
       "      <td>77</td>\n",
       "      <td>8.14</td>\n",
       "      <td>89.54</td>\n",
       "      <td>1.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>79</td>\n",
       "      <td>2.33</td>\n",
       "      <td>91.87</td>\n",
       "      <td>1.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>21</td>\n",
       "      <td>3</td>\n",
       "      <td>82</td>\n",
       "      <td>3.49</td>\n",
       "      <td>95.36</td>\n",
       "      <td>1.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>23</td>\n",
       "      <td>1</td>\n",
       "      <td>83</td>\n",
       "      <td>1.16</td>\n",
       "      <td>96.52</td>\n",
       "      <td>1.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>23</td>\n",
       "      <td>3</td>\n",
       "      <td>86</td>\n",
       "      <td>3.49</td>\n",
       "      <td>100.01</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "   Decile  Number of Cases  Number of Responses  Cumulative Responses  \\\n",
       "0       1               23                   22                    22   \n",
       "1       2               22                   16                    38   \n",
       "2       3               22                   13                    51   \n",
       "3       4               22                   13                    64   \n",
       "4       5               22                    6                    70   \n",
       "5       6               23                    7                    77   \n",
       "6       7               22                    2                    79   \n",
       "7       8               21                    3                    82   \n",
       "8       9               23                    1                    83   \n",
       "9      10               23                    3                    86   \n",
       "\n",
       "   % of Events    Gain  lift  \n",
       "0        25.58   25.58  2.56  \n",
       "1        18.60   44.18  2.21  \n",
       "2        15.12   59.30  1.98  \n",
       "3        15.12   74.42  1.86  \n",
       "4         6.98   81.40  1.63  \n",
       "5         8.14   89.54  1.49  \n",
       "6         2.33   91.87  1.31  \n",
       "7         3.49   95.36  1.19  \n",
       "8         1.16   96.52  1.07  \n",
       "9         3.49  100.01  1.00  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ml.plot_gain_and_lift(rs1,X_test,y_test,pos_label=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Yukarıda ne demiştik, eğer business bizden en güvenilir modeli kurmamızı istiyorsa o zaman accuracy yerine belki log_loss'a bakmayı isteyebiliriz demiştik."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Şimdi direkt en düşük log-loss'u elde etmeyi hedefleyecek şekilde modelimizi kuralım(scoring'e neg_log_loss vereceğiz)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-21T17:32:55.752736Z",
     "start_time": "2022-05-21T17:32:53.590829Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 50 folds for each of 100 candidates, totalling 5000 fits\n",
      "Wall time: 8min 50s\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-4 {color: black;background-color: white;}#sk-container-id-4 pre{padding: 0;}#sk-container-id-4 div.sk-toggleable {background-color: white;}#sk-container-id-4 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-4 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-4 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-4 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-4 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-4 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-4 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-4 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-4 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-4 div.sk-item {position: relative;z-index: 1;}#sk-container-id-4 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-4 div.sk-item::before, #sk-container-id-4 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-4 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-4 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-4 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-4 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-4 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-4 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-4 div.sk-label-container {text-align: center;}#sk-container-id-4 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-4 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-4\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomizedSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                   error_score=&#x27;raise&#x27;,\n",
       "                   estimator=Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                                              FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                                                  kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                                             (&#x27;ct&#x27;,\n",
       "                                              ColumnTransformer(n_jobs=-1,\n",
       "                                                                remainder=&#x27;passthrough&#x27;,\n",
       "                                                                transformers=[(&#x27;nominals&#x27;,\n",
       "                                                                               Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                                                OneHotEncoder(...\n",
       "                                         &#x27;clf__eta0&#x27;: [0.0005, 0.001, 0.002,\n",
       "                                                       0.005],\n",
       "                                         &#x27;clf__l1_ratio&#x27;: [0.9, 1],\n",
       "                                         &#x27;clf__learning_rate&#x27;: [&#x27;constant&#x27;,\n",
       "                                                                &#x27;optimal&#x27;,\n",
       "                                                                &#x27;invscaling&#x27;,\n",
       "                                                                &#x27;adaptive&#x27;],\n",
       "                                         &#x27;clf__penalty&#x27;: [&#x27;l1&#x27;, &#x27;l2&#x27;,\n",
       "                                                          &#x27;elasticnet&#x27;],\n",
       "                                         &#x27;clf__tol&#x27;: [0.0005, 0.001, 0.002,\n",
       "                                                      0.005],\n",
       "                                         &#x27;ct__numerics__ouh&#x27;: [OutlierHandler(featureindices=[0,\n",
       "                                                                                              3]),\n",
       "                                                               None],\n",
       "                                         &#x27;ct__numerics__scl&#x27;: [StandardScaler(),\n",
       "                                                               MinMaxScaler()]}],\n",
       "                   scoring=&#x27;neg_log_loss&#x27;, verbose=1)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-43\" type=\"checkbox\" ><label for=\"sk-estimator-id-43\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomizedSearchCV</label><div class=\"sk-toggleable__content\"><pre>RandomizedSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                   error_score=&#x27;raise&#x27;,\n",
       "                   estimator=Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                                              FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                                                  kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                                             (&#x27;ct&#x27;,\n",
       "                                              ColumnTransformer(n_jobs=-1,\n",
       "                                                                remainder=&#x27;passthrough&#x27;,\n",
       "                                                                transformers=[(&#x27;nominals&#x27;,\n",
       "                                                                               Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                                                OneHotEncoder(...\n",
       "                                         &#x27;clf__eta0&#x27;: [0.0005, 0.001, 0.002,\n",
       "                                                       0.005],\n",
       "                                         &#x27;clf__l1_ratio&#x27;: [0.9, 1],\n",
       "                                         &#x27;clf__learning_rate&#x27;: [&#x27;constant&#x27;,\n",
       "                                                                &#x27;optimal&#x27;,\n",
       "                                                                &#x27;invscaling&#x27;,\n",
       "                                                                &#x27;adaptive&#x27;],\n",
       "                                         &#x27;clf__penalty&#x27;: [&#x27;l1&#x27;, &#x27;l2&#x27;,\n",
       "                                                          &#x27;elasticnet&#x27;],\n",
       "                                         &#x27;clf__tol&#x27;: [0.0005, 0.001, 0.002,\n",
       "                                                      0.005],\n",
       "                                         &#x27;ct__numerics__ouh&#x27;: [OutlierHandler(featureindices=[0,\n",
       "                                                                                              3]),\n",
       "                                                               None],\n",
       "                                         &#x27;ct__numerics__scl&#x27;: [StandardScaler(),\n",
       "                                                               MinMaxScaler()]}],\n",
       "                   scoring=&#x27;neg_log_loss&#x27;, verbose=1)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-44\" type=\"checkbox\" ><label for=\"sk-estimator-id-44\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;log&#x27;,\n",
       "                 FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                                     kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})),\n",
       "                (&#x27;ct&#x27;,\n",
       "                 ColumnTransformer(n_jobs=-1, remainder=&#x27;passthrough&#x27;,\n",
       "                                   transformers=[(&#x27;nominals&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                                   OneHotEncoder(drop=&#x27;first&#x27;,\n",
       "                                                                                 handle_unknown=&#x27;ignore&#x27;))]),\n",
       "                                                  [1, 6, 7]),\n",
       "                                                 (&#x27;numerics&#x27;,\n",
       "                                                  Pipeline(steps=[(&#x27;imp&#x27;,\n",
       "                                                                   SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                                                                  (&#x27;ouh&#x27;,\n",
       "                                                                   DummyTransformer()),\n",
       "                                                                  (&#x27;scl&#x27;,\n",
       "                                                                   DummyTransformer())]),\n",
       "                                                  [2, 3, 4, 5])])),\n",
       "                (&#x27;fs&#x27;,\n",
       "                 SelectKBest(score_func=&lt;function mutual_info_classif at 0x000001BC26E09B80&gt;)),\n",
       "                (&#x27;clf&#x27;, DummyEstimator())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-45\" type=\"checkbox\" ><label for=\"sk-estimator-id-45\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">FunctionTransformer</label><div class=\"sk-toggleable__content\"><pre>FunctionTransformer(func=&lt;function logTransformer at 0x000001BC2B98CE50&gt;,\n",
       "                    kw_args={&#x27;col_name&#x27;: &#x27;Fare&#x27;})</pre></div></div></div><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-46\" type=\"checkbox\" ><label for=\"sk-estimator-id-46\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">ct: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(n_jobs=-1, remainder=&#x27;passthrough&#x27;,\n",
       "                  transformers=[(&#x27;nominals&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;ohe&#x27;,\n",
       "                                                  OneHotEncoder(drop=&#x27;first&#x27;,\n",
       "                                                                handle_unknown=&#x27;ignore&#x27;))]),\n",
       "                                 [1, 6, 7]),\n",
       "                                (&#x27;numerics&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;imp&#x27;,\n",
       "                                                  SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                                                 (&#x27;ouh&#x27;, DummyTransformer()),\n",
       "                                                 (&#x27;scl&#x27;, DummyTransformer())]),\n",
       "                                 [2, 3, 4, 5])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-47\" type=\"checkbox\" ><label for=\"sk-estimator-id-47\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">nominals</label><div class=\"sk-toggleable__content\"><pre>[1, 6, 7]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-48\" type=\"checkbox\" ><label for=\"sk-estimator-id-48\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder(drop=&#x27;first&#x27;, handle_unknown=&#x27;ignore&#x27;)</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-49\" type=\"checkbox\" ><label for=\"sk-estimator-id-49\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">numerics</label><div class=\"sk-toggleable__content\"><pre>[2, 3, 4, 5]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-50\" type=\"checkbox\" ><label for=\"sk-estimator-id-50\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;median&#x27;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-51\" type=\"checkbox\" ><label for=\"sk-estimator-id-51\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyTransformer</label><div class=\"sk-toggleable__content\"><pre>DummyTransformer()</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-52\" type=\"checkbox\" ><label for=\"sk-estimator-id-52\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyTransformer</label><div class=\"sk-toggleable__content\"><pre>DummyTransformer()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-53\" type=\"checkbox\" ><label for=\"sk-estimator-id-53\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">remainder</label><div class=\"sk-toggleable__content\"><pre></pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-54\" type=\"checkbox\" ><label for=\"sk-estimator-id-54\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">passthrough</label><div class=\"sk-toggleable__content\"><pre>passthrough</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-55\" type=\"checkbox\" ><label for=\"sk-estimator-id-55\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SelectKBest</label><div class=\"sk-toggleable__content\"><pre>SelectKBest(score_func=&lt;function mutual_info_classif at 0x000001BC26E09B80&gt;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-56\" type=\"checkbox\" ><label for=\"sk-estimator-id-56\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DummyEstimator</label><div class=\"sk-toggleable__content\"><pre>DummyEstimator()</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomizedSearchCV(cv=RepeatedKFold(n_repeats=10, n_splits=5, random_state=1),\n",
       "                   error_score='raise',\n",
       "                   estimator=Pipeline(steps=[('log',\n",
       "                                              FunctionTransformer(func=<function logTransformer at 0x000001BC2B98CE50>,\n",
       "                                                                  kw_args={'col_name': 'Fare'})),\n",
       "                                             ('ct',\n",
       "                                              ColumnTransformer(n_jobs=-1,\n",
       "                                                                remainder='passthrough',\n",
       "                                                                transformers=[('nominals',\n",
       "                                                                               Pipeline(steps=[('ohe',\n",
       "                                                                                                OneHotEncoder(...\n",
       "                                         'clf__eta0': [0.0005, 0.001, 0.002,\n",
       "                                                       0.005],\n",
       "                                         'clf__l1_ratio': [0.9, 1],\n",
       "                                         'clf__learning_rate': ['constant',\n",
       "                                                                'optimal',\n",
       "                                                                'invscaling',\n",
       "                                                                'adaptive'],\n",
       "                                         'clf__penalty': ['l1', 'l2',\n",
       "                                                          'elasticnet'],\n",
       "                                         'clf__tol': [0.0005, 0.001, 0.002,\n",
       "                                                      0.005],\n",
       "                                         'ct__numerics__ouh': [OutlierHandler(featureindices=[0,\n",
       "                                                                                              3]),\n",
       "                                                               None],\n",
       "                                         'ct__numerics__scl': [StandardScaler(),\n",
       "                                                               MinMaxScaler()]}],\n",
       "                   scoring='neg_log_loss', verbose=1)"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "params= [\n",
    "    {\n",
    "           'clf'               : [SGDClassifier(loss='log_loss',max_iter=mi,random_state=42)],\n",
    "           'clf__alpha'        : [0.005, 0.01, 0.02, 0.05], \n",
    "           'clf__tol'          : [0.0005, 0.001,0.002,0.005],                             \n",
    "           'clf__penalty'      : ['l1','l2','elasticnet'],    \n",
    "           'clf__l1_ratio'     : [0.9, 1],   \n",
    "           'clf__learning_rate': ['constant','optimal','invscaling','adaptive'],\n",
    "           'clf__eta0'          : [0.0005, 0.001, 0.002, 0.005],    \n",
    "           'clf__early_stopping':[True,False], \n",
    "           'clf__class_weight': ['balanced'],                \n",
    "           'ct__numerics__ouh': [OutlierHandler(featureindices=[0,3]),None], #Age ve Fare için, her ne kadar Fare'de LogTrans yapmış olsak da\n",
    "           'ct__numerics__scl' : [StandardScaler(),MinMaxScaler()]\n",
    "          }   \n",
    "] \n",
    "\n",
    "rs_logloss = RandomizedSearchCV(estimator=pipe, param_distributions = params, cv = mycv, n_jobs=-1, verbose = 1, \n",
    "                          scoring = 'neg_log_loss',error_score='raise',n_iter=100)\n",
    "rs_logloss.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-05-21T17:33:02.443968Z",
     "start_time": "2022-05-21T17:33:02.403084Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div id=388217c2-bb2f-418b-82b0-12a48963a26d style=\"display:none; background-color:#9D6CFF; color:white; width:200px; height:30px; padding-left:5px; border-radius:4px; flex-direction:row; justify-content:space-around; align-items:center;\" onmouseover=\"this.style.backgroundColor='#BA9BF8'\" onmouseout=\"this.style.backgroundColor='#9D6CFF'\" onclick=\"window.commands?.execute('create-mitosheet-from-dataframe-output');\">See Full Dataframe in Mito</div> <script> if (window.commands?.hasCommand('create-mitosheet-from-dataframe-output')) document.getElementById('388217c2-bb2f-418b-82b0-12a48963a26d').style.display = 'flex' </script> <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>param_ct__numerics__scl</th>\n",
       "      <th>param_ct__numerics__ouh</th>\n",
       "      <th>param_clf__tol</th>\n",
       "      <th>param_clf__penalty</th>\n",
       "      <th>param_clf__learning_rate</th>\n",
       "      <th>param_clf__l1_ratio</th>\n",
       "      <th>param_clf__eta0</th>\n",
       "      <th>param_clf__early_stopping</th>\n",
       "      <th>param_clf__class_weight</th>\n",
       "      <th>param_clf__alpha</th>\n",
       "      <th>param_clf</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>StandardScaler()</td>\n",
       "      <td>OutlierHandler(featureindices=[0, 3])</td>\n",
       "      <td>0.001</td>\n",
       "      <td>elasticnet</td>\n",
       "      <td>optimal</td>\n",
       "      <td>0.9</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>False</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.005</td>\n",
       "      <td>SGDClassifier(alpha=0.005, class_weight='balanced', eta0=0.0005, l1_ratio=0.9,\\n              loss='log_loss', max_iter=4000, penalty='elasticnet',\\n              random_state=42)</td>\n",
       "      <td>-0.455551</td>\n",
       "      <td>0.045144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>StandardScaler()</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>elasticnet</td>\n",
       "      <td>optimal</td>\n",
       "      <td>1</td>\n",
       "      <td>0.001</td>\n",
       "      <td>False</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.005</td>\n",
       "      <td>SGDClassifier(alpha=0.005, class_weight='balanced', eta0=0.0005, l1_ratio=0.9,\\n              loss='log_loss', max_iter=4000, penalty='elasticnet',\\n              random_state=42)</td>\n",
       "      <td>-0.458206</td>\n",
       "      <td>0.042674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>MinMaxScaler()</td>\n",
       "      <td>None</td>\n",
       "      <td>0.001</td>\n",
       "      <td>elasticnet</td>\n",
       "      <td>optimal</td>\n",
       "      <td>0.9</td>\n",
       "      <td>0.001</td>\n",
       "      <td>False</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.005</td>\n",
       "      <td>SGDClassifier(alpha=0.005, class_weight='balanced', eta0=0.0005, l1_ratio=0.9,\\n              loss='log_loss', max_iter=4000, penalty='elasticnet',\\n              random_state=42)</td>\n",
       "      <td>-0.461107</td>\n",
       "      <td>0.038702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>StandardScaler()</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>elasticnet</td>\n",
       "      <td>optimal</td>\n",
       "      <td>0.9</td>\n",
       "      <td>0.001</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.01</td>\n",
       "      <td>SGDClassifier(alpha=0.005, class_weight='balanced', eta0=0.0005, l1_ratio=0.9,\\n              loss='log_loss', max_iter=4000, penalty='elasticnet',\\n              random_state=42)</td>\n",
       "      <td>-0.461324</td>\n",
       "      <td>0.042655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>StandardScaler()</td>\n",
       "      <td>None</td>\n",
       "      <td>0.0005</td>\n",
       "      <td>l1</td>\n",
       "      <td>constant</td>\n",
       "      <td>0.9</td>\n",
       "      <td>0.005</td>\n",
       "      <td>False</td>\n",
       "      <td>balanced</td>\n",
       "      <td>0.005</td>\n",
       "      <td>SGDClassifier(alpha=0.005, class_weight='balanced', eta0=0.0005, l1_ratio=0.9,\\n              loss='log_loss', max_iter=4000, penalty='elasticnet',\\n              random_state=42)</td>\n",
       "      <td>-0.464388</td>\n",
       "      <td>0.036489</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table></div>"
      ],
      "text/plain": [
       "   param_ct__numerics__scl                param_ct__numerics__ouh  \\\n",
       "99        StandardScaler()  OutlierHandler(featureindices=[0, 3])   \n",
       "3         StandardScaler()                                   None   \n",
       "40          MinMaxScaler()                                   None   \n",
       "38        StandardScaler()                                   None   \n",
       "51        StandardScaler()                                   None   \n",
       "\n",
       "   param_clf__tol param_clf__penalty param_clf__learning_rate  \\\n",
       "99          0.001         elasticnet                  optimal   \n",
       "3          0.0005         elasticnet                  optimal   \n",
       "40          0.001         elasticnet                  optimal   \n",
       "38         0.0005         elasticnet                  optimal   \n",
       "51         0.0005                 l1                 constant   \n",
       "\n",
       "   param_clf__l1_ratio param_clf__eta0 param_clf__early_stopping  \\\n",
       "99                 0.9          0.0005                     False   \n",
       "3                    1           0.001                     False   \n",
       "40                 0.9           0.001                     False   \n",
       "38                 0.9           0.001                      True   \n",
       "51                 0.9           0.005                     False   \n",
       "\n",
       "   param_clf__class_weight param_clf__alpha  \\\n",
       "99                balanced            0.005   \n",
       "3                 balanced            0.005   \n",
       "40                balanced            0.005   \n",
       "38                balanced             0.01   \n",
       "51                balanced            0.005   \n",
       "\n",
       "                                            param_clf  mean_test_score  \\\n",
       "99  SGDClassifier(alpha=0.005, class_weight='balan...        -0.455551   \n",
       "3   SGDClassifier(alpha=0.005, class_weight='balan...        -0.458206   \n",
       "40  SGDClassifier(alpha=0.005, class_weight='balan...        -0.461107   \n",
       "38  SGDClassifier(alpha=0.005, class_weight='balan...        -0.461324   \n",
       "51  SGDClassifier(alpha=0.005, class_weight='balan...        -0.464388   \n",
       "\n",
       "    std_test_score  \n",
       "99        0.045144  \n",
       "3         0.042674  \n",
       "40        0.038702  \n",
       "38        0.042655  \n",
       "51        0.036489  "
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ml.gridsearch_to_df(rs_logloss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.45555076906923936"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rs_logloss.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.757847533632287"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(y_test,rs_logloss.predict(X_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-2.67998516, -0.41291619,  0.78841002, -1.08395218, -1.06679707,\n",
       "       -3.51209754, -2.93858746, -0.98334107,  3.99902193, -0.52928144])"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rs1.best_estimator_[\"clf\"].coef_[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# ml.linear_model_feature_importance(rs1,coltrans,\"fs\",\"clf\")  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is still not implemented and not planned as it seems out of scope of sklearn, as per Github discussion #6773 and #13048.\n",
    "\n",
    "However, the documentation on linear models now mention that (P-value estimation note):\n",
    "\n",
    "It is theoretically possible to get p-values and confidence intervals for coefficients in cases of regression without penalization.\n",
    "The statsmodels package natively supports this.\n",
    "Within sklearn, one could use bootstrapping.\n",
    "\n",
    "\n",
    "- LinReg için burda implemntasyon var: https://gist.github.com/brentp/5355925\n",
    "- LogReg için :https://gist.github.com/rspeare/77061e6e317896be29c6de9a85db301d"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from scipy.stats import norm\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "def logit_pvalue(model, x):\n",
    "    \"\"\" Calculate z-scores for scikit-learn LogisticRegression.\n",
    "    parameters:\n",
    "        model: fitted sklearn.linear_model.LogisticRegression with intercept and large C\n",
    "        x:     matrix on which the model was fit\n",
    "    This function uses asymtptics for maximum likelihood estimates.\n",
    "    \"\"\"\n",
    "    p = model.predict_proba(x)\n",
    "    n = len(p)\n",
    "    m = len(model.coef_[0]) + 1\n",
    "    coefs = np.concatenate([model.intercept_, model.coef_[0]])\n",
    "    x_full = np.matrix(np.insert(np.array(x), 0, 1, axis = 1))\n",
    "    ans = np.zeros((m, m))\n",
    "    for i in range(n):\n",
    "        ans = ans + np.dot(np.transpose(x_full[i, :]), x_full[i, :]) * p[i,1] * p[i, 0]\n",
    "    vcov = np.linalg.inv(np.matrix(ans))\n",
    "    se = np.sqrt(np.diag(vcov))\n",
    "    t =  coefs/se  \n",
    "    p = (1 - norm.cdf(abs(t))) * 2\n",
    "    return p\n",
    "\n",
    "# test p-values\n",
    "x = np.arange(10)[:, np.newaxis]\n",
    "y = np.array([0,0,0,1,0,0,1,1,1,1])\n",
    "model = LogisticRegression(C=1e30).fit(x, y)\n",
    "print(logit_pvalue(model, x))\n",
    "\n",
    "# compare with statsmodels\n",
    "import statsmodels.api as sm\n",
    "sm_model = sm.Logit(y, sm.add_constant(x)).fit(disp=0)\n",
    "print(sm_model.pvalues)\n",
    "sm_model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Kaynaklar"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html\n",
    "- https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.SGDClassifier.html \n",
    "- https://scikit-learn.org/stable/modules/sgd.html\n",
    "- https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.RidgeClassifier.html \n",
    "- https://towardsdatascience.com/logistic-regression-detailed-overview-46c4da4303bc\n",
    "- https://medium.com/analytics-vidhya/your-guide-for-logistic-regression-with-titanic-dataset-784943523994\n",
    "- https://www.kaggle.com/code/mnassrib/titanic-logistic-regression-with-python/notebook\n",
    "- https://www.analyticsvidhya.com/blog/2021/07/titanic-survival-prediction-using-machine-learning/\n",
    "- https://becominghuman.ai/titanic-survival-dataset-part-2-2-logistic-regression-7ebe9e30bf54"
   ]
  }
 ],
 "metadata": {
  "finalized": {
   "timestamp": 1657623638864,
   "trusted": true
  },
  "hide_input": false,
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "671.818px",
    "left": "347px",
    "top": "176.051px",
    "width": "304.474px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}