{ "cells": [ { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "#pip install pycaret" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "1.0.0\n" ] } ], "source": [ "from pycaret.utils import version\n", "version()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# 1. Loading dataset" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
agesexbmichildrensmokerregioncharges
019female27.9000yessouthwest16884.92400
118male33.7701nosoutheast1725.55230
228male33.0003nosoutheast4449.46200
333male22.7050nonorthwest21984.47061
432male28.8800nonorthwest3866.85520
\n", "
" ], "text/plain": [ " age sex bmi children smoker region charges\n", "0 19 female 27.900 0 yes southwest 16884.92400\n", "1 18 male 33.770 1 no southeast 1725.55230\n", "2 28 male 33.000 3 no southeast 4449.46200\n", "3 33 male 22.705 0 no northwest 21984.47061\n", "4 32 male 28.880 0 no northwest 3866.85520" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from pycaret.datasets import get_data\n", "data = get_data('insurance')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# 2. Initializing Setup" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " \n", "Setup Succesfully Completed!\n" ] }, { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Description Value
0session_id123
1Transform Target False
2Transform Target MethodNone
3Original Data(1338, 7)
4Missing Values False
5Numeric Features 2
6Categorical Features 4
7Ordinal Features False
8High Cardinality Features False
9High Cardinality Method None
10Sampled Data(1338, 7)
11Transformed Train Set(936, 61)
12Transformed Test Set(402, 61)
13Numeric Imputer mean
14Categorical Imputer constant
15Normalize True
16Normalize Method zscore
17Transformation False
18Transformation Method None
19PCA False
20PCA Method None
21PCA Components None
22Ignore Low Variance False
23Combine Rare Levels False
24Rare Level Threshold None
25Numeric Binning True
26Remove Outliers False
27Outliers Threshold None
28Remove Multicollinearity False
29Multicollinearity Threshold None
30Clustering False
31Clustering Iteration None
32Polynomial Features True
33Polynomial Degree 2
34Trignometry Features True
35Polynomial Threshold 0.100000
36Group Features False
37Feature Selection False
38Features Selection Threshold None
39Feature Interaction True
40Feature Ratio False
41Interaction Threshold 0.010000
" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from pycaret.regression import *\n", "\n", "reg1 = setup(data, target = 'charges', session_id = 123,\n", " normalize = True,\n", " polynomial_features = True, trigonometry_features = True, feature_interaction=True, \n", " bin_numeric_features= ['age', 'bmi'])" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
MAEMSERMSER2RMSLEMAPE
02815.84312.048964e+074526.54780.87760.39400.2995
13607.36373.481924e+075900.78280.79750.45550.3529
23021.02072.326287e+074823.16010.70550.58150.3948
33193.79662.591245e+075090.42720.78360.54320.4356
43172.60502.788106e+075280.25170.79260.45160.2867
53222.10652.310720e+074806.99450.85140.37860.2861
62865.43982.438587e+074938.20510.83410.39300.3251
73400.18892.942705e+075424.67050.83820.47640.3169
83003.74022.194569e+074684.62320.85830.38330.3208
93154.98442.829284e+075319.10160.81780.49870.3544
Mean3145.70892.595239e+075079.47640.81570.45560.3373
SD225.81154.044394e+06388.98520.04670.06690.0458
\n", "
" ], "text/plain": [ " MAE MSE RMSE R2 RMSLE MAPE\n", "0 2815.8431 2.048964e+07 4526.5478 0.8776 0.3940 0.2995\n", "1 3607.3637 3.481924e+07 5900.7828 0.7975 0.4555 0.3529\n", "2 3021.0207 2.326287e+07 4823.1601 0.7055 0.5815 0.3948\n", "3 3193.7966 2.591245e+07 5090.4272 0.7836 0.5432 0.4356\n", "4 3172.6050 2.788106e+07 5280.2517 0.7926 0.4516 0.2867\n", "5 3222.1065 2.310720e+07 4806.9945 0.8514 0.3786 0.2861\n", "6 2865.4398 2.438587e+07 4938.2051 0.8341 0.3930 0.3251\n", "7 3400.1889 2.942705e+07 5424.6705 0.8382 0.4764 0.3169\n", "8 3003.7402 2.194569e+07 4684.6232 0.8583 0.3833 0.3208\n", "9 3154.9844 2.829284e+07 5319.1016 0.8178 0.4987 0.3544\n", "Mean 3145.7089 2.595239e+07 5079.4764 0.8157 0.4556 0.3373\n", "SD 225.8115 4.044394e+06 388.9852 0.0467 0.0669 0.0458" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "lr = create_model('lr')" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plot_model(lr)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# 3. Save Model for deployment" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Transformation Pipeline and Model Succesfully Saved\n" ] } ], "source": [ "save_model(lr, 'deployment_28042020')" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Transformation Pipeline and Model Sucessfully Loaded\n" ] } ], "source": [ "deployment_28042020 = load_model('deployment_28042020')" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[Pipeline(memory=None,\n", " steps=[('dtypes',\n", " DataTypes_Auto_infer(categorical_features=[],\n", " display_types=True, features_todrop=[],\n", " ml_usecase='regression',\n", " numerical_features=[], target='charges',\n", " time_features=[])),\n", " ('imputer',\n", " Simple_Imputer(categorical_strategy='not_available',\n", " numeric_strategy='mean',\n", " target_variable=None)),\n", " ('new_levels1',\n", " New_Catagorical_Levels...\n", " ('dummy', Dummify(target='charges')),\n", " ('fix_perfect', Remove_100(target='charges')),\n", " ('clean_names', Clean_Colum_Names()),\n", " ('feature_select', Empty()), ('fix_multi', Empty()),\n", " ('dfs',\n", " DFS_Classic(interactions=['multiply'], ml_usecase='regression',\n", " random_state=123, subclass='binary',\n", " target='charges',\n", " top_features_to_pick_percentage=None)),\n", " ('pca', Empty())],\n", " verbose=False),\n", " LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False),\n", " None]" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "deployment_28042020" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# 4. Generate Predictions from deployed app" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "75714.0\n" ] } ], "source": [ "import requests\n", "url = 'https://pycaret-demo1.herokuapp.com/predict_api'\n", "pred = requests.post(url,json={'age':55, 'sex':'male', 'bmi':59, 'children':1, 'smoker':'male', 'region':'northwest'})\n", "print(pred.json())" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "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": "3.7.4" } }, "nbformat": 4, "nbformat_minor": 2 }