{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **Use ML techniques for layer stackup modeling**  \n",
    "## Table of contents:  \n",
    "### 1.**[Motivation](#Motivation)**  <a href='#Motivation'>...</a>\n",
    "### 2.**[Problem Statements](#Problem_Statements)**  <a href='#Problem_Statements'>...</a>\n",
    "### 3.**[Generate Data](#Generate_Data)**  <a href='#Generate_Data'>...</a>\n",
    "### 4.**[Prepare Data](#Prepare_Data)**  <a href='#Prepare_Data'>...</a>\n",
    "### 5.**[Choose a Model](#Choose_a_Model)**  <a href='#Choose_a_Model'>...</a>\n",
    "### 6.**[Training](#Training)**  <a href='#Training'>...</a>\n",
    "### 7.**[Neural Network](#Neural_Network)**  <a href='#Neural_Network'>...</a>\n",
    "### 8.**[Deploy](#Deploy)**  <a href='#Deploy'>...</a>\n",
    "### 9.**[Conclusion](#Conclusion)**  <a href='#Conclusion'>...</a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Motivation:**<a id = 'Motivation'></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When planning a PCB layer stackup, often time we would like to know the trade-off between various layout options vs their signal integrity performance. For example, a wider trace may provides smaller impedance yet occupate more routing area. Narrow down the spacing between differential pairs may save some spaces but will also increase crosstalk. Most EDA tool involving system level signal integrity analysis provides \"transmission line calculator\" like shown below for designer to quickly make estimation and determine the trade-off:  \n",
    "![TLineCalc](https://github.com/SPISim/ML_LStkModeling/blob/master/assets/images/KiCad_LineCalc.png?raw=true)\n",
    "\n",
    "However, all such \"calculators\" I have seen, even in a commercial one, only consider the single trace or one differential pair itself. They do not take take crosstalks into account. More over, stackup parameters such as conductivity and permetivity must be entered individually instead of a range. As a results, user can't not easily visualize relationships between performance parameters vs the stackup properties. Thus, an enhanced version of such \"T-Line calculator\", which can address the aformentioned gaps will be very useful. Such tool requires a prediction model to link between various stackup parameters to their performance targets. Data science/machine learning techniques can thus be used to build such model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Problem Statements:**<a id = 'Problem_Statements'></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We would like to build a prediction model such that given a set of stackup parameters such as trace width and spacing etc, its performance such as impedance, attenuations, near-end and far-end crosstalk can be quickly estimated. This model can then be deployed into a stand-alone tool for range based sweep such that a visual plot can be generated to provide relations between various parameters to decide design trade-off."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Generate Data**:<a id = 'Generate_Data'></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Overview:   \n",
    "The model to be built here is for nominal (i.e. numerical) prediction with around 10 attributes, i.e. input variables. Various stackup configurations will be generated via sampling and their corresponding stackup model, in the form of frequency dependent R/L/G/C matrices will be simulated via field solver. Such process are deterministics. Post process steps will read these solved model and calculate performance. Here we define performance to be predicted as impedance, attenuation, near-end/far-end crosstalks and propagation speed."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ***Define stakup structure:***   \n",
    "There are many possible stakup structures as shown below. For more accurate prediction, we are going to generate one prediction model per structure.\n",
    "![Presets](https://github.com/SPISim/ML_LStkModeling/blob/master/assets/images/LStkPresets.png?raw=true)\n",
    "\n",
    "Use three single-ended traces (victim in the middle) in strip-line setup as an example, various attributes may be defined as shown below:\n",
    "![Setup](https://github.com/SPISim/ML_LStkModeling/blob/master/assets/images/SLSE3Setup.png?raw=true)\n",
    "\n",
    "These parameters, such as S(Spacing), W(Width), Sigma(Conductivity), Er(Permitivity), H(Height) etc are represented as varaibles to be sampled."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ***Define sampling points:***   \n",
    "Next step is to define ranges of variable values and sampling points. Since there are about 10 parameters, full combinatorial data will be impractical. Thus we may need to apply sampling algorithms such as design-of-experiments or spacing filling etc to establish best coverage of the solution space.\n",
    "For this setup, we have generate 10,000 cases to be simulated.\n",
    "![Sample](https://github.com/SPISim/ML_LStkModeling/blob/master/assets/images/SLSE3Sample.png?raw=true)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ***Generate inputs setup and simulate:***   \n",
    "Once we have sample points, layer stackup configurations to the solver to be used will be generated. Each field solver has different syntax thus a flow will be needed...\n",
    "![TProFlow](https://github.com/SPISim/ML_LStkModeling/blob/master/assets/images/SLSE3FlowGUI.png?raw=true)\n",
    "\n",
    "In this case, we use HSpice from Synopsys for field solver, thus each of 10K parameter combinations will be used to generate their spice input files for simulation:\n",
    "![Simulate](https://github.com/SPISim/ML_LStkModeling/blob/master/assets/images/SLSE3Simulate.png?raw=true)\n",
    "\n",
    "The next step is to perforam circuit simulation for all these cases. This may be a time-consuming process so a distributed environment or simulation farm may be used."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ***Performance measurement:***   \n",
    "The outcome of each simulation is a frequency dependent tabular model, corresponding to its layer stackup settings. HSpice's tabular format looks like this:\n",
    "![Tabular](https://github.com/SPISim/ML_LStkModeling/blob/master/assets/images/SLSE3Tabular.png?raw=true)\n",
    "Next step is to load these models and do performance measurement:\n",
    "![Measure](https://github.com/SPISim/ML_LStkModeling/blob/master/assets/images/SLSE3Measure.png?raw=true)\n",
    "Matrix manipulation such as eigen-value decomponsition will be applied in order to obtain the characteristic impedance and propagation speed etc. Measurement output of each model should be a set of parameters which will be combined with original inputs to form the dataset for our prediction modeling."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Prepare Data:**<a id = 'Prepare_Data'></a>\n",
    "From this point, we can start the modeling process using python and various packages."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "## Initial set-up for data\n",
    "import os\n",
    "import pandas as pd\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "prjHome = 'C:/Temp/WinProj/LStkMdl'\n",
    "workDir = prjHome + '/wsp/'\n",
    "srcFile = prjHome + '/dat/SLSE3.csv'\n",
    "\n",
    "def save_fig(fig_id, tight_layout=True, fig_extension=\"png\", resolution=300):\n",
    "    path = os.path.join(workDir, fig_id + \".\" + fig_extension)\n",
    "    print(\"Saving figure\", fig_id)\n",
    "    if tight_layout:\n",
    "        plt.tight_layout()\n",
    "    plt.savefig(path, format=fig_extension, dpi=resolution)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>H1</th>\n",
       "      <th>H3</th>\n",
       "      <th>ER1</th>\n",
       "      <th>ER3</th>\n",
       "      <th>TD1</th>\n",
       "      <th>TD3</th>\n",
       "      <th>H2</th>\n",
       "      <th>W</th>\n",
       "      <th>S</th>\n",
       "      <th>EDW</th>\n",
       "      <th>SIGMA</th>\n",
       "      <th>H</th>\n",
       "      <th>FNAME</th>\n",
       "      <th>Z0SE(1_SE)</th>\n",
       "      <th>S0SE(1_SE)</th>\n",
       "      <th>KBSENB(1_1)</th>\n",
       "      <th>KFSENB(1_1)</th>\n",
       "      <th>A(1_1)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4.9311</td>\n",
       "      <td>5.5859</td>\n",
       "      <td>3.6964</td>\n",
       "      <td>3.3277</td>\n",
       "      <td>0.013771</td>\n",
       "      <td>0.030565</td>\n",
       "      <td>0.91236</td>\n",
       "      <td>7.5573</td>\n",
       "      <td>30.6610</td>\n",
       "      <td>0.177530</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15</td>\n",
       "      <td>SPIMDL00001.TAB</td>\n",
       "      <td>37.715088</td>\n",
       "      <td>1.601068e+08</td>\n",
       "      <td>0.000035</td>\n",
       "      <td>-7.297762e-15</td>\n",
       "      <td>0.497289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.8918</td>\n",
       "      <td>5.3156</td>\n",
       "      <td>4.0410</td>\n",
       "      <td>3.8404</td>\n",
       "      <td>0.022008</td>\n",
       "      <td>0.008600</td>\n",
       "      <td>0.53126</td>\n",
       "      <td>7.0378</td>\n",
       "      <td>6.9217</td>\n",
       "      <td>0.733370</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15</td>\n",
       "      <td>SPIMDL00002.TAB</td>\n",
       "      <td>39.831954</td>\n",
       "      <td>1.510345e+08</td>\n",
       "      <td>0.019608</td>\n",
       "      <td>-1.093612e-12</td>\n",
       "      <td>0.249591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.7907</td>\n",
       "      <td>11.5230</td>\n",
       "      <td>3.6984</td>\n",
       "      <td>3.3534</td>\n",
       "      <td>0.013859</td>\n",
       "      <td>0.045137</td>\n",
       "      <td>2.00990</td>\n",
       "      <td>3.6964</td>\n",
       "      <td>27.3230</td>\n",
       "      <td>0.707750</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15</td>\n",
       "      <td>SPIMDL00003.TAB</td>\n",
       "      <td>35.663928</td>\n",
       "      <td>1.587798e+08</td>\n",
       "      <td>0.000305</td>\n",
       "      <td>-2.797417e-13</td>\n",
       "      <td>0.705953</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2.8595</td>\n",
       "      <td>4.7259</td>\n",
       "      <td>3.9605</td>\n",
       "      <td>3.3981</td>\n",
       "      <td>0.010481</td>\n",
       "      <td>0.018028</td>\n",
       "      <td>0.36547</td>\n",
       "      <td>2.3872</td>\n",
       "      <td>7.2803</td>\n",
       "      <td>0.735210</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15</td>\n",
       "      <td>SPIMDL00004.TAB</td>\n",
       "      <td>59.456438</td>\n",
       "      <td>1.558444e+08</td>\n",
       "      <td>0.010125</td>\n",
       "      <td>-5.478920e-12</td>\n",
       "      <td>0.327952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.5946</td>\n",
       "      <td>8.2553</td>\n",
       "      <td>3.4176</td>\n",
       "      <td>3.3249</td>\n",
       "      <td>0.024434</td>\n",
       "      <td>0.003663</td>\n",
       "      <td>2.08810</td>\n",
       "      <td>6.3776</td>\n",
       "      <td>20.1680</td>\n",
       "      <td>0.053585</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15</td>\n",
       "      <td>SPIMDL00005.TAB</td>\n",
       "      <td>46.082722</td>\n",
       "      <td>1.633106e+08</td>\n",
       "      <td>0.004136</td>\n",
       "      <td>-4.787903e-13</td>\n",
       "      <td>0.169760</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       H1       H3     ER1     ER3       TD1       TD3       H2       W  \\\n",
       "0  4.9311   5.5859  3.6964  3.3277  0.013771  0.030565  0.91236  7.5573   \n",
       "1  4.8918   5.3156  4.0410  3.8404  0.022008  0.008600  0.53126  7.0378   \n",
       "2  1.7907  11.5230  3.6984  3.3534  0.013859  0.045137  2.00990  3.6964   \n",
       "3  2.8595   4.7259  3.9605  3.3981  0.010481  0.018028  0.36547  2.3872   \n",
       "4  5.5946   8.2553  3.4176  3.3249  0.024434  0.003663  2.08810  6.3776   \n",
       "\n",
       "         S       EDW       SIGMA   H            FNAME  Z0SE(1_SE)  \\\n",
       "0  30.6610  0.177530  50000000.0  15  SPIMDL00001.TAB   37.715088   \n",
       "1   6.9217  0.733370  50000000.0  15  SPIMDL00002.TAB   39.831954   \n",
       "2  27.3230  0.707750  50000000.0  15  SPIMDL00003.TAB   35.663928   \n",
       "3   7.2803  0.735210  50000000.0  15  SPIMDL00004.TAB   59.456438   \n",
       "4  20.1680  0.053585  50000000.0  15  SPIMDL00005.TAB   46.082722   \n",
       "\n",
       "     S0SE(1_SE)  KBSENB(1_1)   KFSENB(1_1)    A(1_1)  \n",
       "0  1.601068e+08     0.000035 -7.297762e-15  0.497289  \n",
       "1  1.510345e+08     0.019608 -1.093612e-12  0.249591  \n",
       "2  1.587798e+08     0.000305 -2.797417e-13  0.705953  \n",
       "3  1.558444e+08     0.010125 -5.478920e-12  0.327952  \n",
       "4  1.633106e+08     0.004136 -4.787903e-13  0.169760  "
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's read the data and do some statistic\n",
    "srcData = pd.read_csv(srcFile)\n",
    "\n",
    "# Take a peek:\n",
    "srcData.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 10000 entries, 0 to 9999\n",
      "Data columns (total 18 columns):\n",
      "H1             10000 non-null float64\n",
      "H3             10000 non-null float64\n",
      "ER1            10000 non-null float64\n",
      "ER3            10000 non-null float64\n",
      "TD1            10000 non-null float64\n",
      "TD3            10000 non-null float64\n",
      "H2             10000 non-null float64\n",
      "W              10000 non-null float64\n",
      "S              10000 non-null float64\n",
      "EDW            10000 non-null float64\n",
      "SIGMA          10000 non-null float64\n",
      "H              10000 non-null int64\n",
      "FNAME          10000 non-null object\n",
      "Z0SE(1_SE)     10000 non-null float64\n",
      "S0SE(1_SE)     10000 non-null float64\n",
      "KBSENB(1_1)    9998 non-null float64\n",
      "KFSENB(1_1)    9998 non-null float64\n",
      "A(1_1)         10000 non-null float64\n",
      "dtypes: float64(16), int64(1), object(1)\n",
      "memory usage: 1.4+ MB\n"
     ]
    }
   ],
   "source": [
    "srcData.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>H1</th>\n",
       "      <th>H3</th>\n",
       "      <th>ER1</th>\n",
       "      <th>ER3</th>\n",
       "      <th>TD1</th>\n",
       "      <th>TD3</th>\n",
       "      <th>H2</th>\n",
       "      <th>W</th>\n",
       "      <th>S</th>\n",
       "      <th>EDW</th>\n",
       "      <th>SIGMA</th>\n",
       "      <th>H</th>\n",
       "      <th>Z0SE(1_SE)</th>\n",
       "      <th>S0SE(1_SE)</th>\n",
       "      <th>KBSENB(1_1)</th>\n",
       "      <th>KFSENB(1_1)</th>\n",
       "      <th>A(1_1)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>1.000000e+04</td>\n",
       "      <td>9.998000e+03</td>\n",
       "      <td>9.998000e+03</td>\n",
       "      <td>10000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.199999</td>\n",
       "      <td>7.500007</td>\n",
       "      <td>3.850001</td>\n",
       "      <td>3.850001</td>\n",
       "      <td>0.025000</td>\n",
       "      <td>0.025000</td>\n",
       "      <td>1.400000</td>\n",
       "      <td>5.500001</td>\n",
       "      <td>18.500007</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>42.985345</td>\n",
       "      <td>1.533140e+08</td>\n",
       "      <td>1.475118e-02</td>\n",
       "      <td>-2.155868e-12</td>\n",
       "      <td>0.509534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.616661</td>\n",
       "      <td>3.175591</td>\n",
       "      <td>0.490773</td>\n",
       "      <td>0.490772</td>\n",
       "      <td>0.014434</td>\n",
       "      <td>0.014434</td>\n",
       "      <td>0.635117</td>\n",
       "      <td>2.020829</td>\n",
       "      <td>9.526750</td>\n",
       "      <td>0.216515</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>172.163835</td>\n",
       "      <td>7.085288e+06</td>\n",
       "      <td>2.847164e-02</td>\n",
       "      <td>3.787910e-11</td>\n",
       "      <td>0.288974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.400000</td>\n",
       "      <td>2.001000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000100</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.300140</td>\n",
       "      <td>2.000100</td>\n",
       "      <td>2.002600</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>11.630424</td>\n",
       "      <td>1.384625e+08</td>\n",
       "      <td>1.009146e-10</td>\n",
       "      <td>-2.796522e-09</td>\n",
       "      <td>0.011486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2.799925</td>\n",
       "      <td>4.750325</td>\n",
       "      <td>3.425075</td>\n",
       "      <td>3.425050</td>\n",
       "      <td>0.012500</td>\n",
       "      <td>0.012501</td>\n",
       "      <td>0.850038</td>\n",
       "      <td>3.750350</td>\n",
       "      <td>10.250000</td>\n",
       "      <td>0.187510</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>32.173933</td>\n",
       "      <td>1.479284e+08</td>\n",
       "      <td>2.216767e-04</td>\n",
       "      <td>-1.112024e-12</td>\n",
       "      <td>0.328938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>4.200000</td>\n",
       "      <td>7.500500</td>\n",
       "      <td>3.850000</td>\n",
       "      <td>3.849950</td>\n",
       "      <td>0.025000</td>\n",
       "      <td>0.025000</td>\n",
       "      <td>1.399950</td>\n",
       "      <td>5.500150</td>\n",
       "      <td>18.500000</td>\n",
       "      <td>0.374985</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>40.031614</td>\n",
       "      <td>1.527750e+08</td>\n",
       "      <td>1.969018e-03</td>\n",
       "      <td>-1.928585e-15</td>\n",
       "      <td>0.507237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>5.600075</td>\n",
       "      <td>10.249500</td>\n",
       "      <td>4.274950</td>\n",
       "      <td>4.274950</td>\n",
       "      <td>0.037499</td>\n",
       "      <td>0.037498</td>\n",
       "      <td>1.949875</td>\n",
       "      <td>7.249650</td>\n",
       "      <td>26.748000</td>\n",
       "      <td>0.562507</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>48.688036</td>\n",
       "      <td>1.583456e+08</td>\n",
       "      <td>1.376321e-02</td>\n",
       "      <td>7.447876e-13</td>\n",
       "      <td>0.680500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6.999900</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>4.699800</td>\n",
       "      <td>4.699900</td>\n",
       "      <td>0.050000</td>\n",
       "      <td>0.049995</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>8.999600</td>\n",
       "      <td>34.999000</td>\n",
       "      <td>0.749930</td>\n",
       "      <td>50000000.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>17097.973570</td>\n",
       "      <td>1.725066e+08</td>\n",
       "      <td>3.184933e-01</td>\n",
       "      <td>9.613918e-11</td>\n",
       "      <td>14.524624</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 H1            H3           ER1           ER3           TD1  \\\n",
       "count  10000.000000  10000.000000  10000.000000  10000.000000  10000.000000   \n",
       "mean       4.199999      7.500007      3.850001      3.850001      0.025000   \n",
       "std        1.616661      3.175591      0.490773      0.490772      0.014434   \n",
       "min        1.400000      2.001000      3.000000      3.000100      0.000003   \n",
       "25%        2.799925      4.750325      3.425075      3.425050      0.012500   \n",
       "50%        4.200000      7.500500      3.850000      3.849950      0.025000   \n",
       "75%        5.600075     10.249500      4.274950      4.274950      0.037499   \n",
       "max        6.999900     13.000000      4.699800      4.699900      0.050000   \n",
       "\n",
       "                TD3            H2             W             S           EDW  \\\n",
       "count  10000.000000  10000.000000  10000.000000  10000.000000  10000.000000   \n",
       "mean       0.025000      1.400000      5.500001     18.500007      0.375000   \n",
       "std        0.014434      0.635117      2.020829      9.526750      0.216515   \n",
       "min        0.000002      0.300140      2.000100      2.002600      0.000002   \n",
       "25%        0.012501      0.850038      3.750350     10.250000      0.187510   \n",
       "50%        0.025000      1.399950      5.500150     18.500000      0.374985   \n",
       "75%        0.037498      1.949875      7.249650     26.748000      0.562507   \n",
       "max        0.049995      2.500000      8.999600     34.999000      0.749930   \n",
       "\n",
       "            SIGMA        H    Z0SE(1_SE)    S0SE(1_SE)   KBSENB(1_1)  \\\n",
       "count     10000.0  10000.0  10000.000000  1.000000e+04  9.998000e+03   \n",
       "mean   50000000.0     15.0     42.985345  1.533140e+08  1.475118e-02   \n",
       "std           0.0      0.0    172.163835  7.085288e+06  2.847164e-02   \n",
       "min    50000000.0     15.0     11.630424  1.384625e+08  1.009146e-10   \n",
       "25%    50000000.0     15.0     32.173933  1.479284e+08  2.216767e-04   \n",
       "50%    50000000.0     15.0     40.031614  1.527750e+08  1.969018e-03   \n",
       "75%    50000000.0     15.0     48.688036  1.583456e+08  1.376321e-02   \n",
       "max    50000000.0     15.0  17097.973570  1.725066e+08  3.184933e-01   \n",
       "\n",
       "        KFSENB(1_1)        A(1_1)  \n",
       "count  9.998000e+03  10000.000000  \n",
       "mean  -2.155868e-12      0.509534  \n",
       "std    3.787910e-11      0.288974  \n",
       "min   -2.796522e-09      0.011486  \n",
       "25%   -1.112024e-12      0.328938  \n",
       "50%   -1.928585e-15      0.507237  \n",
       "75%    7.447876e-13      0.680500  \n",
       "max    9.613918e-11     14.524624  "
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "srcData.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ** Note that:**  \n",
    "- Sigma(Conductivity) and H(default layer height) are constants in this setup;\n",
    "- FNAME (File name) is not needed for modeling\n",
    "- Z0 (impedance) has outliers\n",
    "- Forward/Backward crosstalk (Kb/kf) have missing terms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop constant and file name columns\n",
    "stkData = srcData.drop(columns=['H', 'SIGMA', 'FNAME'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure attribute_histogram_plots\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1fa03108048>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot distributions before dropping measurement outliers\n",
    "stkData.hist(bins=50, figsize=(20,15))\n",
    "save_fig(\"attribute_histogram_plots\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 9994 entries, 0 to 9999\n",
      "Data columns (total 15 columns):\n",
      "H1             9994 non-null float64\n",
      "H3             9994 non-null float64\n",
      "ER1            9994 non-null float64\n",
      "ER3            9994 non-null float64\n",
      "TD1            9994 non-null float64\n",
      "TD3            9994 non-null float64\n",
      "H2             9994 non-null float64\n",
      "W              9994 non-null float64\n",
      "S              9994 non-null float64\n",
      "EDW            9994 non-null float64\n",
      "Z0SE(1_SE)     9994 non-null float64\n",
      "S0SE(1_SE)     9994 non-null float64\n",
      "KBSENB(1_1)    9994 non-null float64\n",
      "KFSENB(1_1)    9994 non-null float64\n",
      "A(1_1)         9994 non-null float64\n",
      "dtypes: float64(15)\n",
      "memory usage: 1.2 MB\n"
     ]
    }
   ],
   "source": [
    "# drop outliers and invalid Kb/Kf cells\n",
    "# These may be caused by unphysical stakup model or calculation during post-processing\n",
    "maxZVal = 200\n",
    "minZVal = 10\n",
    "stkTemp = stkData[(stkData['Z0SE(1_SE)'] < maxZVal) & \\\n",
    "                  (stkData['Z0SE(1_SE)'] > minZVal) & \\\n",
    "                  (np.abs(stkData['KBSENB(1_1)']) > 0.0) & \\\n",
    "                  (np.abs(stkData['KFSENB(1_1)']) > 0.0)]\n",
    "\n",
    "# Check again to make sure data are now justified\n",
    "stkTemp.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure attribute_histogram_plots\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1fa033cf2e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# now plot distributions again, should see proper distribuition now\n",
    "stkData = stkTemp\n",
    "stkData.hist(bins=50, figsize=(20,15))\n",
    "save_fig(\"attribute_histogram_plots\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Z0SE(1_SE)    1.000000\n",
       "W             0.664396\n",
       "H1            0.535432\n",
       "H3            0.344356\n",
       "H2            0.186618\n",
       "ER1           0.119390\n",
       "ER3           0.110507\n",
       "EDW           0.067261\n",
       "TD3           0.061906\n",
       "TD1           0.017937\n",
       "S             0.000107\n",
       "Name: Z0SE(1_SE), dtype: float64"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find principal components for Z\n",
    "corr_matrix = stkData.drop(columns=['KBSENB(1_1)', 'KFSENB(1_1)', 'S0SE(1_SE)', 'A(1_1)']).corr()\n",
    "corr_matrix['Z0SE(1_SE)'].abs().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From this correlation matrix above, it can be shown that trace width and height are dominate factors for the trace's impedance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Choose a Model:**<a id = 'Choose_a_Model'></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since we are building a nominal estimator here, I will try simple linear regressor as estimator first:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Separate input and output attributes\n",
    "allTars = ['Z0SE(1_SE)', 'KBSENB(1_1)', 'KFSENB(1_1)', 'S0SE(1_SE)', 'A(1_1)']\n",
    "varList = [e for e in list(stkData) if e not in allTars]\n",
    "varData = stkData[varList]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.478480854776983"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We have 10,000 cases here, try in-memory normal equation directly first:\n",
    "\n",
    "# LinearRegression Fit Impedance\n",
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "tarData = stkData['Z0SE(1_SE)']\n",
    "lin_reg = LinearRegression()\n",
    "lin_reg.fit(varData, tarData)\n",
    "\n",
    "# Fit and check predictions using MSE etc\n",
    "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
    "predict = lin_reg.predict(varData)\n",
    "resRMSE = np.sqrt(mean_squared_error(tarData, predict))\n",
    "resRMSE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Attribute: ['H1', 'H3', 'ER1', 'ER3', 'TD1', 'TD3', 'H2', 'W', 'S', 'EDW']\n",
      "Scores: [3.25291562 4.16881619 3.22429058 3.32251943 3.70096691 3.46812853\n",
      " 3.34169206 3.5511994  3.28259776 3.39876687]\n",
      "Mean: 3.4711893352193854\n",
      "Standard deviation: 0.270957271481177\n"
     ]
    }
   ],
   "source": [
    "# Use 10-Split for cross validations:\n",
    "def display_scores(attribs, scores):\n",
    "    print(\"Attribute:\", attribs)\n",
    "    print(\"Scores:\", scores)\n",
    "    print(\"Mean:\", scores.mean())\n",
    "    print(\"Standard deviation:\", scores.std())\n",
    "    \n",
    "from sklearn.model_selection import cross_val_score\n",
    "lin_scores = cross_val_score(lin_reg, varData, tarData,\n",
    "                             scoring=\"neg_mean_squared_error\", cv=10)\n",
    "lin_rmse_scores = np.sqrt(-lin_scores)\n",
    "display_scores(varList, lin_rmse_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.4921877809274595"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# try Regularization it self\n",
    "from sklearn.linear_model import Ridge\n",
    "ridge_reg = Ridge(alpha=1, solver=\"cholesky\")\n",
    "ridge_reg.fit(varData, tarData)\n",
    "predict = ridge_reg.predict(varData)\n",
    "resRMSE = np.sqrt(mean_squared_error(tarData, predict))\n",
    "resRMSE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Thus a 3 ohms or so difference may be obtained from this estimator. What if higher order regression is used:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.2780298094097002"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import PolynomialFeatures\n",
    "poly_features = PolynomialFeatures(degree=2, include_bias=False)\n",
    "varPoly = poly_features.fit_transform(varData)\n",
    "lin_reg = LinearRegression()\n",
    "lin_reg.fit(varPoly, tarData)\n",
    "predict = lin_reg.predict(varPoly)\n",
    "resRMSE = np.sqrt(mean_squared_error(tarData, predict))\n",
    "resRMSE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A more accurate model thus may be obtained this way. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Training and Evaluation:**<a id = 'Training'></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure underfitting_learning_curves_plot\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzt3Xl8FeXZ//HPBYQ9IsiiIgIq+AhqQQJiUcEFRVxwa0WxgoqoP1vBqhXUR8W6PnV7LFalglAXtNi6o5jHuhQXIBFQIApCUYHIouKGIoHr98dM4kk4CSfhLJOT7/v1mldmuWfmOmFyLuaee+7b3B0REZGoqZfpAEREROJRghIRkUhSghIRkUhSghIRkUhSghIRkUhSghIRkUhKW4Iysw5m9pqZFZnZIjMbHa5vZWb5ZrY0/Nmykv2Hh2WWmtnwdMUtIiKZYel6D8rMdgN2c/f3zCwXKAROBkYAX7r7bWY2Fmjp7ldV2LcVUADkAR7u28vdv0pL8CIiknZpu4Ny92J3fy+c/xYoAtoDQ4CpYbGpBEmromOBfHf/MkxK+cCg1EctIiKZ0iATJzWzTkBPYDbQzt2LIUhiZtY2zi7tgc9illeG6+IdexQwKljapRd0olevZEUuIiJVKSwsXO/ubZJxrLQnKDNrDvwDGOPu35hZQrvFWRe3btLdJwITg3PlORRQUFDTaEVEpDrM7JNkHSutrfjMLIcgOT3m7v8MV68Jn0+VPqdaG2fXlUCHmOU9gNWpjFVERDIrna34DJgEFLn7XTGbngNKW+UNB56Ns/tM4Bgzaxm28jsmXCciIlkqnXdQ/YDfAEea2fxwGgzcBgw0s6XAwHAZM8szs4cA3P1L4I/A3HC6MVwnIiJZKm3NzDOh9BlUFn9EEZFIMbNCd89LxrHUk4SIiERSRpqZi0hmfPPNN6xdu5bNmzdnOhSphXJycmjbti077bRTWs6nBCVSR3zzzTesWbOG9u3b06RJExJ8xUMEAHfnhx9+YNWqVQBpSVKq4hOpI9auXUv79u1p2rSpkpNUm5nRtGlT2rdvz9q18d4GSj4lKJE6YvPmzTRp0iTTYUgt16RJk7RVEStBidQhunOSHZXOa0gJSkREIkkJSkREIkkJSkTqrKFDh3L66adXa5++fftyxRVXpCgiiaVm5iISWdt73jF8+HCmTJlS4+M/+OCDVLc3nRkzZpCTk1Pjc6bL0KFDKSkp4amnnsp0KDWmBCUiNXPDDcGUQsXFxWXzL7zwAhdccEG5dZW1Sty8eXNCSaRFixbVjqlVq1bV3kdqRlV8IlIz48en/BS77rpr2bTzzjtvs65FixZ8+OGHmBnTp0+nf//+NG7cmKlTp7JmzRrOOOOMsne/9t9/fx577LFyx69Yxde3b18uu+wyrrzySlq1asWuu+7K1VdfXe4uq2IV36677srtt9/OeeedR25uLh06dODee+8td57FixfTr18/GjduTLdu3cjPz6dBgwY88cQTlX72efPmMWDAAHJzc8nNzaVnz57MmjWrbPsHH3zAoEGDaN68Oe3atePss89m3bp1AIwdO5Ynn3ySf/zjH5gZZsa7775bg3+BzFKCEqnLzGo+7cj+KTB27Fguu+wyioqKGDx4MD/88AN9+/blxRdfZOHChVx88cUMHz683Jd8PJMnT6ZFixbMnj2bO++8k9tvv51nnnmmyn3uuOMO+vTpw7x58xg9ejSjR4/mvffeA6CkpIQhQ4aQm5vLnDlzmDhxIldffTVbt26t8pi//vWv6dy5MwUFBcybN49rr72WRo0aAfDZZ59x+OGH07t3bwoLC5k5cybr16/n1FNPBeDaa69lyJAhnHDCCRQXF1NcXEyv2ji0uLtn7QS9HFxE3H3x4sXbroTMTDUwffp0J86+RUVFDviECRO2e4whQ4b4JZdcUrZ8xhln+GmnnVa2fPDBB/uAAQPK7XPooYeW2+fggw/2yy+/vGy5Xbt2PmLEiHL77LHHHv6nP/3J3d2feeYZz8nJ8TVr1pRtf/XVVx3wadOmxY1z69at3qhRI3/iiSfibr/yyit98ODB5dYVFxc74AsWLIj72ZIp7rUUAgo8Sd/huoMSqct2JM3syP4pkJdXfoSHkpISxo8fzwEHHECrVq1o3rw5L774Ip9++mmVxznwwAPLLe++++7b7dqnqn0+/PBDOnXqRNu2bcu2H3zwwVUez8y47LLLOPvssxk4cCC33norS5cuLdteWFhIfn4+zZs3L5v22WcfAJYtW1blsWsTJSgRyQrNmjUrt3zzzTdz3333MW7cOF577TXmz5/P4MGD+emnn6o8TsXGFWa23eq4qvZx9xr1vnDrrbeycOFCBg8ezJtvvkn37t3LnqFt3bqVk08+mfnz55ebli5dysCBA6t9rqhSKz4RqZnrr890BFWaNWsWp5xyCmeddRYQfKkvWbKEjh07pjWO/fbbjxUrVrBu3TratGkDwJw5cxLad99992Xfffflsssu49xzz2XSpEkMGzaMgw46iJdffpnOnTtTv379uPs2bNiQTZs2Je1zZILuoESkZlLcxHxHde3alZkzZ/LOO+9QVFTEhRdeyOrVq9Mex/HHH8+ee+7J8OHDef/993nrrbcYO3ZsWeu6eL7++msuvfRS3njjDT755BPefvtt3nnnHbp16wbA6NGjKS4u5qyzzmLu3LksX76cV155hfPPP7/sDrFTp04sWLCApUuXsn79ekpKStL2mZNFCUpEstL48eM58MADGThwIAMGDKBt27bV7jUiGRo0aMCzzz7Lhg0b6N27NyNHjuS6664DoHHjxnH3ycnJYe3atfzmN7+ha9eu/OpXv+KII47g9ttvB2DPPffk7bffZtOmTQwcOJD999+fSy+9lObNm5fdUV188cV07tyZnj170qZNGwoKCtLzgZPIPEUPLKPALM+hIFXPZEVqlaKiIvbbb79MhyHA7Nmz6du3LwsXLqR79+6ZDqfaqrqWzKzQ3fPibqymtD2DMrPJwAnAWnffP1z3JLBvWGRnYIO794iz7wrgW2ALUJKsDy8ikg7Tp0+nZcuW7LPPPixbtowxY8bQp0+fWpmc0imdjSSmABOAv5WucPczSufN7E7g6yr2P8Ld16csOhGRFPn6668ZN24cK1euZJddduGoo47izjvvzHRYkZe2BOXub5pZp3jbLHhS+GvgyHTFIyKSLiNHjmTkyJGZDqPWiUojicOANe6+tJLtDrxiZoVmNqqqA5nZKDMrMLPa90RQRETKROU9qDOBaVVs7+fuq82sLZBvZh+6+5vxCrr7RGAilDaSEBGR2ijjd1Bm1gA4FXiysjLuvjr8uRZ4GuiTnuhERCRTMp6ggKOBD919ZbyNZtbMzHJL54FjgIVpjE9ERDIgbQnKzKYB7wD7mtlKMzs/3DSUCtV7Zra7mc0IF9sBs8xsATAHeNHdX05X3CIikhnpbMV3ZiXrR8RZtxoYHM4vB36R0uBERCRyolDFJyKScg899FDZqLzxluO57bbbyoaxSOa5JTFKUCISWSeeeCJHH3103G1FRUWYGfn5+TU69rBhw1iyZMmOhLeNkpISzGybEXhTca5UOPvsszn55JMzHUYZJSgRiayRI0fyr3/9ixUrVmyzbdKkSXTs2JGjjjqqRsdu0qRJuUEEUymd58omSlAiElnHH3887dq14+GHHy63fvPmzTzyyCOcd9551KsXfI1dccUVdO3alSZNmtC5c2fGjh1b5XhI8ardbr31Vtq1a0dubi4jRoxg48aN5bbPnj2bgQMH0rp1a1q0aMFhhx1WbmynTp06AXDKKadgZmXVg/HO9Ze//IW9996bhg0b0qVLFyZPnly2rfRO7KGHHuK0006jWbNm7L333kybVtXrorBgwQKOPPJIdtppJ3Jzc+nRowdvvPFG2faFCxdy3HHHkZubS9u2bRk2bBhr1qwB4Nprr+Wxxx7j2WefLRsKZNasWVWeL9WUoETqMLPMTYlo0KABw4cPZ8qUKeVGtX3++edZv3495557btm6nXbaiSlTplBUVMSECRN49NFHue222xL+XTz++OPccMMN3HTTTRQWFrLXXntxzz33lCvz7bffMnz4cP7973/z7rvvcsABB3Dcccfx1VdfATB37lwAHn74YYqLi3n33Xfjnmv69OmMGTOGyy+/nIULF3LJJZcwatQoXnrppXLlxo8fz2mnncaCBQs49dRTGTFiBCtXxn0jB4ChQ4fSoUMH5syZw7x587juuuvKhvRYtWoV/fv3p2fPnsydO5f8/Hw2bNjAKaecgrszduxYTjvtNAYNGkRxcTHFxcXbHZo+5dw9ayfo5eAi4u6LFy/eZh1kbkrUkiVLHPCZM2eWrRs8eLAPGjSoyv3+/Oc/+7777lu2/Ne//tVbtGhR6XLv3r39oosuKneM/v37+957713pObZu3eqtW7f2adOmubv75s2bHfCnn366XLmK5+rTp49fcMEF5coMGzbM+/fvX+441157bdn2TZs2ecOGDcvOFU/Tpk390Ucfjbtt3Lhxfswxx5Rbt27dOge8sLCwLIYhQ4ZUevxS8a6lUkCBJ+k7XHdQIhJpXbp04fDDDy+rAlu9ejUzZ87cpvPVJ598kn79+rHrrrvSvHlzrrjiCj799NOEz1NUVMQhhxxSbl3F5TVr1jBq1Ci6du1KixYtyM3N5YsvvqjWeUrP1a9fv3LrDj30UBYvXlxu3YEHHlg237BhQ1q3bs3atWsrPe7vf/97RowYwdFHH80tt9xSrmFGYWEhr732Gs2bNy+bSqskly1bVq3400UJSqQOy+Q9VHWMHDmSZ555hi+//JIpU6bQqlUrTjrppLLts2bNYtiwYQwePJjnn3+eefPmceONN5YNf54sZ599NvPmzeOee+7h7bffZv78+ey+++41Ok+84d4rrsvJydlme2xVZ0V//OMfWbRoESeccAKzZs1i//33Z+rUqQBs3bqVE088kfnz55ebli5dynHHHVft+NNBCUpEIu/000+ncePGPProo0yePJlzzjmn3Jf3W2+9RceOHbnmmmvo3bs3Xbp0idvyryr77bffNs+MKi7PmjWLSy+9lMGDB9O9e3eaNWvG559/Xra9fv361K9fny1btmz3XBUbIMyaNYtu3bpVK+Z4unbtypgxY5gxYwbDhw9n0qRJABx00EEsWrSITp06sc8++5SbmjdvDgR3aduLPZ2UoEQk8po0acJZZ53FDTfcwLJlyzj//PPLbe/atSuffvop06ZNY9myZUyYMIG///3v1TrH6NGjmTx5MpMmTWLJkiVljSUqnueRRx6hqKiIOXPmMHToUBo1alS23czYc889efXVV/n888/LGk9UdOWVVzJlyhTuv/9+li5dyj333MMTTzzBH/7wh2rFHOu7777jd7/7HW+88QaffPIJ7777Lm+99VZZ0vvd737H+vXrOfPMM5kzZw7Lly8nPz+fkSNH8sMPPwBBK8T333+fJUuWsH79ekpKSmocTzIoQYlIrTBy5Ei++uorfvnLX7LffvuV23bKKadw2WWXcemll9KjRw9ef/11xo8fX63jDxs2jGuvvZZx48Zx0EEH8eGHHzJ69OhyZaZMmcKGDRvo2bMnZ511FhdeeCEdOnQoV+auu+4iPz+fDh060Lt377jnOv3007nnnnu444476N69O/fddx8PPvjgDlW1NWjQgPXr13POOefQtWtXTj31VA477DDuuOMOAPbYYw/eeusttmzZwrHHHkv37t357W9/S9OmTcvuRi+88EK6dOlCr169aNOmTaWtENPFvLqVwbVIMB5UQbXru0WyUVFR0TZf7CI1UdW1ZGaF7p6XjPPoDkpERCJJCUpERCJJCUpERCJJCUpERCJJCUqkDsnmRlGSHum8hpSgROqInJycsvddRGrqhx9+2KaHi1RRghKpI9q2bcuqVavYuHGj7qSk2tydjRs3smrVqrSNbdUgLWcBzGwycAKw1t33D9fdAFwArAuLXe3uM+LsOwj4X6A+8JC7J96HvogAwXAUEHS2unnz5gxHI7VRTk4O7dq1K7uWUi1tCQqYAkwA/lZh/d3ufkdlO5lZfeA+YCCwEphrZs+5++LK9hGR+Hbaaae0fbmI7Ki0VfG5+5vAlzXYtQ/wsbsvd/efgCeAIUkNTkREIicKz6B+a2bvm9lkM2sZZ3t74LOY5ZXhurjMbJSZFZhZQbIDFRGR9Ml0grof2BvoARQDd8YpE29w6Eqf8Lr7RHfPS1ZfUCIikhkZTVDuvsbdt7j7VuCvBNV5Fa0EYrsL3gNYnY74REQkczKaoMxst5jFU4CFcYrNBbqYWWczawgMBZ5LR3wiIpI56WxmPg0YALQ2s5XA9cAAM+tBUGW3ArgwLLs7QXPywe5eYma/BWYSNDOf7O6L0hW3iIhkhsaDEhGRpNF4UCIikvWUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJKUoEREJJLSlqDMbLKZrTWzhTHr/mRmH5rZ+2b2tJntXMm+K8zsAzObb2YF6YpZREQyJ513UFOAQRXW5QP7u/uBwBJgXBX7H+HuPdw9L0XxiYhIhKQtQbn7m8CXFda94u4l4eK7wB7pikdERKItSs+gzgNeqmSbA6+YWaGZjUpjTCIikiENMh0AgJldA5QAj1VSpJ+7rzaztkC+mX0Y3pHFO9YoIExivVIQrYiIpEPG76DMbDhwAjDM3T1eGXdfHf5cCzwN9KnseO4+0d3z9KxKRKR2y2iCMrNBwFXASe6+sZIyzcwst3QeOAZYGK+siIhkj3Q2M58GvAPsa2Yrzex8YAKQS1BtN9/MHgjL7m5mM8Jd2wGzzGwBMAd40d1fTlfcIiKSGVZJrVpWMMtzKCCLP6KISKSYWWGyHrFk/BmUiIhIPEpQIiISSUpQIiISSUpQIiISSUpQIiISSUpQIiISSUpQIiISSUpQIiISSQklKDO7xcyaxiwPNrMmMcs7mdnfUhGgiIjUTYneQV0FNI9ZfgLYLWa5CTAsWUGJiIgkmqBsO8siIiJJpWdQIiISSUpQIiISSdUZUfciM/suZr/zzeyLcDk3uWGJiEhdl2iC+hQ4N2b5c+CsOGVERESSIqEE5e6dUhyHiIhIOXoGJSIikZToi7q/MLMjKqwbZmbLzWytmT1gZg1TE6KIiNRFid5B3QQcWrpgZt2Ah4GlwDSCl3SvSnp0IiJSZyWaoA4C8mOWhwKL3f1Ydx8NjAHOSHZwIiJSdyWaoHYBVsUsHw48H7P8OrDn9g5iZpPDKsGFMetamVm+mS0Nf7asZN/hYZmlZjY8wbhFRKSWSjRBrQPaA5hZfaAXMDtme0NgawLHmQIMqrBuLPCqu3cBXg2XyzGzVsD1wMFAH+D6yhKZiIhkh0QT1OsESWEv4PJw3Wsx27sBK7Z3EHd/E/iywuohwNRwfipwcpxdjwXy3f1Ld/+KoLqxYqITEZEskuiLuv8N/B/wMbAFuNTdv4/Z/huCu5+aaOfuxQDuXmxmbeOUaQ98FrO8Mly3DTMbBYwKlnrVMCQREcm0RF/UXWFm/wV0B9a5++oKRa4nSBqpEq/3dI9X0N0nAhMBzPLilhERkehL+EVddy9x9wVxkhPh+i/i7ZeANWa2G0D4c22cMiuBDjHLewDbxCEiItkjoTsoM/t9IuXc/a4axPAcMBy4Lfz5bJwyM4FbYhpGHAOMq8G5RESklkj0GdQdwHrgOyofrNCBKhOUmU0DBgCtzWwlQdXgbcDfzex8gg5nfxWWzQMucveR7v6lmf0RmBse6kZ3r9jYQkREsoi5b/8xjZnNIWipNx2Y5O6zUh1YMgTPoApI4COKiEgSmFmhu+cl41gJPYNy9z4E7yB9BfzTzD4ysz+YWbtkBCEiIlJRdRpJLHL33xM0776GoKpuhZk9a2aNUhSfiIjUUdUZURcAd98MPGVm3wBNgeOBJsCmJMcmIiJ1WLXGgzKzTmZ2o5l9AvwV+DfQxd03pCQ6ERGpsxJtZn4WcD5wCEEnsRcCMz2RFhYiIiI1kGgV36METcDvIWhu3g3oZla+xXkN34MSERHZRqIJ6lOC95zOrKLMdt+DEhERSVSiffF12l4ZM+uwvTIiIiKJqlYjiXjMbFczmwAsSUI8IiIiQIIJysx2NrPHzGydma02s0stcD2wnOAl3vNSGqmIiNQpiT6DuoVgmPepBAMF3g0MBJoBx7n7G6kJT0RE6qpEE9TxwLnu/n9m9heCgQuXufuY1IUmIiJ1WaLPoHYHFgO4+3LgR4IXdUVERFIi0QRVD9gcs7wF2Jj8cERERAKJVvEZ8KiZlfa31xj4q5mVS1LuflIygxMRkbor0QQ1tcLyo8kOREREJFaiL+qem+pAREREYu3wi7pRd/rpmY5ARERqIusTVJMmmY5ARERqIuMJysz2NbP5MdM3ZjamQpkBZvZ1TJnrMhWviIikR7VH1E02d/8I6AFgZvWBVcDTcYr+291PSGdsIiKSORm/g6rgKIIeKj5J1gE1pKKISO0UtQQ1FJhWybZDzGyBmb1kZt3TGZSIiKRfZBKUmTUETgKmx9n8HtDR3X8B/Bl4porjjDKzAjMrSE2kIiKSDpFJUMBxwHvuvqbiBnf/xt2/C+dnADlm1jreQdx9orvnuXteasMVEZFUilKCOpNKqvfCQREtnO9DEPcXaYxNRETSLOOt+ADMrCnB+FIXxqy7CMDdHwBOBy42sxLgB2Cou5o/iIhks0gkKHffCOxSYd0DMfMTgAnpjktERDInSlV8KaH7LBGR2inrE9Ty5ZmOQEREaiLrE9Q772Q6AhERqYmsT1AiIlI7KUGJiEgkKUGJiEgkKUGJiEgkZX2CahCJN71ERKS6sj5B9e6d6QhERKQmsj5BlZRkOgIREamJrE9Qc+dmOgIREamJrE9QIiJSOylBiYhIJGV9gurcOdMRiIhITWR9gtq8OdMRiIhITWR9glq5MtMRiIhITWR9ghIRkdqpTiSorVszHYGIiFRXnUhQCxdmOgIREamuOpGgxozJdAQiIlJdkUlQZrbCzD4ws/lmVhBnu5nZvWb2sZm9b2YHJXrsn35KbqwiIpJ6Uevr+wh3X1/JtuOALuF0MHB/+HO73JMTnIiIpE9k7qASMAT4mwfeBXY2s90yHZSIiKRGlBKUA6+YWaGZjYqzvT3wWczyynBdOWY2yswKYqsJdQclIlL7RKmKr5+7rzaztkC+mX3o7m/GbLc4+2yTetx9IjARwCzPAerXT0W4IiKSSpG5g3L31eHPtcDTQJ8KRVYCHWKW9wBWJ3LsQw5JRoQiIpJOkUhQZtbMzHJL54FjgIpvLz0HnBO25usLfO3uxYkcv3nzpIYrIiJpEJUqvnbA02YGQUyPu/vLZnYRgLs/AMwABgMfAxuBcxM9uDqMFRGpfSKRoNx9OfCLOOsfiJl34JKaHF/vQYmI1D6RqOJLNd1BiYjUPnUiQa2v7NVfERGJrDqRoD77bPtlREQkWupEgnr99UxHICIi1VUnEhTA1KmZjkBERKqjziSoESMyHYGIiFRHnUlQAFu2ZDoCSTmz2j/tvDPccEOmf5MiGZfVCapRo/LLAwfuwMFiv0DqZfWvrfb68cdMR5AcX38N48dnPlHWxqlBg+qVv+GGqrfVxI5+hk6dgine+eP9ZyaLmWdxV989euT5ggXlxz78n/+BK69M8ABmyQlkR3/HpRfj1q0/L1c8brJi3VEdO8KKFek/73/+AwccAN9/n/5zi0iZPKDAPSlfSFmdoPLy8vzHHwtYtGjbbZdcAtw3gTwKOOmXX9Do7X/RjI1pj1FEJJsoQSUoLy/P584tqHaNXEM2cTT/x6HMojXr+YqWDOB1WvIVnfkP9dkSd+wPEZG6LpkJKhJ98aWSGZSUBFXTifqJRszgeGZwfPxjspVd+IIm/MCufM5mcthEIwxnJ74BYCNNqcdW6rGVBpRQny38REN+oiEN+YlGbCr7WXFqxvdl+9ZjKzlsxrYd+irjPMJpOqqxOZZQbJX9e1d1HSR7W9SPV9NzpUoyz5msY2Xme+O2pB0p6xMUQP01q3HaM5NjOJHn2UzDHTqeU4/1tAHgM/ZMRogiIlmhlxJUNRx2GMyaBcCxvMJPNMKBT+hIBz6jPlv5kUYUsR9fsAvz6MliutGSr9hIU97hEBpQwmZyWMkefEWrzH4eEZE6IrufQZl5wfaLVdtP5PAFu/AdzVlHGxqxicb8yBbq8x3NcYxmfI9jbKE+W6hPCQ1oxCZy2FxW1bdt5V4jfqQx39OsrCpoK/X4aQfv+FIpilWP3HwLEJ2GjRWVNsqsTGV/klX9qSZ7W9SPt822P98bNM8HvFFjuCKBpro33ww4WD24+uo422/a/jFKY6lYbXv1NQnvC8AtNwMGjRrim7YzPlCHPWHlSvCt1YtpBziW8N96PlerkUQikpag+veveYd+6fqWbNQoGu8BZTor7Mi/lYjsMDMrdPe8ZBwr+6v4tsf95//S7rknfPLJjr+3VPH4dUld+7wikjJ1O0G1aBH81JeqiEjk1M0+e0rvmjZsyHQkIiJSiYwnKDPrYGavmVmRmS0ys9Fxygwws6/NbH44XVejkzVq9PPd0taqHzCKiEhmRaGKrwS43N3fM7NcoNDM8t19cYVy/3b3E2p8FlXjiYjUKhm/g3L3Ynd/L5z/FigC2mc2KhERybSMJ6hYZtYJ6AnMjrP5EDNbYGYvmVn3Ko4xyswKzOznFua6exIRqXUik6DMrDnwD2CMu39TYfN7QEd3/wXwZ+CZyo7j7hPdPS9Z7fBFRCQzIpGgzCyHIDk95u7/rLjd3b9x9+/C+RlAjpm1TnOYIiKSRhlPUGZmwCSgyN3vqqTMrmE5zKwPQdxfVOMkwTRgwI4HLCIiaRGFVnz9gN8AH5jZ/HDd1RB0E+7uDwCnAxebWQnwAzDUs7mPJhERyXyCcvdZUHWvhu4+AZiQnohERCQKMp6g0kI3WyIitU7Gn0GlhZ4/iYjUOnUjQQHLdg5dAAAL3klEQVSsWJHpCEREpBrqToLq1CnTEYiISDVk/zOoZctgr70yHYWIiFRT9t9B1cv+jygiko2y/9tbCUpEpFbK/m9vJSgRkVop+7+9rcp3gEVEJKKyP0HpDkpEpFbK/m9vJSgRkVop+7+9VcUnIlIrZX+C0h2UiEitlP3f3m3aqC8+EZFaKPt7kvjyS2jZMtNRiIhINWX/HZSq+EREaqXs//ZWghIRqZWy/9tbrfhERGql7E9QuoMSEamVsv/bWwlKRKRWisS3t5kNMrOPzOxjMxsbZ3sjM3sy3D7bzDpV4+BJjFRERNIl4wnKzOoD9wHHAd2AM82sW4Vi5wNfufs+wN3A7QmfQHdQIiK1UhS+vfsAH7v7cnf/CXgCGFKhzBBgajj/FHCUWYK3RkpQIiK1UhRe1G0PfBazvBI4uLIy7l5iZl8DuwDrKx7MzEYBowgL5DX4+SMWQmEyA0+y1sT5PBGmeFNL8aZObYoVal+8+ybrQFFIUPHuhLwGZYKV7hOBiQBmVrDePW/HwksPMyvwWhIrKN5UU7ypU5tihdoZb7KOFYX6r5VAh5jlPYDVlZUxswZAC+DLtEQnIiIZEYUENRfoYmadzawhMBR4rkKZ54Dh4fzpwL/cPe4dlIiIZIeMV/GFz5R+C8wE6gOT3X2Rmd0IFLj7c8Ak4BEz+5jgzmlogoefmJKgU6M2xQqKN9UUb+rUplihDsdruhEREZEoikIVn4iIyDaUoEREJJKyMkFtr+ukNMYx2czWmtnCmHWtzCzfzJaGP1uG683M7g1jft/MDorZZ3hYfqmZDY93riTE2sHMXjOzIjNbZGajIx5vYzObY2YLwnjHh+s7h91hLQ27x2oYrq+0uywzGxeu/8jMjk1FvDHnqm9m88zshajHa2YrzOwDM5tf2nQ4qtdDeJ6dzewpM/swvI4PiWq8ZrZv+Hstnb4xszERjvey8O9soZlNC//+Un/tuntWTQQNLZYBewENgQVAtwzFcjhwELAwZt3/AGPD+bHA7eH8YOAlgne++gKzw/WtgOXhz5bhfMsUxLobcFA4nwssIeh6KqrxGtA8nM8BZodx/B0YGq5/ALg4nP9/wAPh/FDgyXC+W3iNNAI6h9dO/RReE78HHgdeCJcjGy+wAmhdYV0kr4fwXFOBkeF8Q2DnKMcbE3d94HOgYxTjJego4T9Ak5hrdkQ6rt2U/dIzNQGHADNjlscB4zIYTyfKJ6iPgN3C+d2Aj8L5B4EzK5YDzgQejFlfrlwK434WGFgb4gWaAu8R9ECyHmhQ8VogaCV6SDjfICxnFa+P2HIpiHMP4FXgSOCF8PxRjncF2yaoSF4PwE4EX6JWG+KtEOMxwFtRjZefe/JpFV6LLwDHpuPazcYqvnhdJ7XPUCzxtHP3YoDwZ9twfWVxp/3zhLfkPQnuSiIbb1hdNh9YC+QT/I9sg7uXxDl3ue6ygNLustL5+70H+AOwNVzeJeLxOvCKmRVa0IUYRPd62AtYBzwcVqE+ZGbNIhxvrKHAtHA+cvG6+yrgDuBToJjgWiwkDdduNiaohLtFipjK4k7r5zGz5sA/gDHu/k1VReOsS2u87r7F3XsQ3Jn0Afar4twZjdfMTgDWuntsf5BVnTvjv1+gn7sfRDDSwCVmdngVZTMdbwOC6vT73b0n8D1BFVllMh1vEETw3OYkYPr2isZZl5Z4w+dgQwiq5XYHmhFcE5WdN2mxZmOCSqTrpExaY2a7AYQ/14brK4s7bZ/HzHIIktNj7v7PqMdbyt03AK8T1M3vbEF3WBXPXVl3WemKtx9wkpmtIOix/0iCO6qoxou7rw5/rgWeJvhPQFSvh5XASnefHS4/RZCwohpvqeOA99x9TbgcxXiPBv7j7uvcfTPwT+CXpOHazcYElUjXSZkU223TcIJnPaXrzwlb6/QFvg5v8WcCx5hZy/B/MseE65LKzIygx44id7+rFsTbxsx2DuebEPwRFQGvEXSHFS/eeN1lPQcMDVsedQa6AHOSHa+7j3P3Pdy9E8E1+S93HxbVeM2smZnlls4T/DsuJKLXg7t/DnxmZqU9aR8FLI5qvDHO5OfqvdK4ohbvp0BfM2safk+U/m5Tf+2m8uFfpiaCFi9LCJ5JXJPBOKYR1NluJvjfw/kEdbGvAkvDn63CskYwcOMy4AMgL+Y45wEfh9O5KYr1UILb7feB+eE0OMLxHgjMC+NdCFwXrt8rvOg/Jqg2aRSubxwufxxu3yvmWNeEn+Mj4Lg0XBcD+LkVXyTjDeNaEE6LSv+Oono9hOfpARSE18QzBK3aohxvU+ALoEXMukjGC4wHPgz/1h4haImX8mtXXR2JiEgkZWMVn4iIZAElKBERiSQlKBERiSQlKBERiSQlKBERiSQlKJEYZvaEmT1VzX3eNbM7UhVTlJjZf5mZm9n+mY5Fsp+amUutYmbbu2CnuvuIHTh+C4K/iw3V2KcVsNndv63pedPBzJ4g6Nzz9O0WrvwY9YE2wHr/uR82kZRosP0iIpGyW8z8CcBfK6z7Id5OZpbjQTctVXL3r6sbkLt/Wd19ait330IwNIRIyqmKT2oVd/+8dAI2VFzn7l/HVEP9yszeMLMfgeFm1i4cSG2VmW0MB18bFnv8ilV8YfXd3Wb2JzP70sw+N7Nbwi5fYsvcEbP8uZldZcGAld+a2WdmdmmF83Qzs7fM7EczW2xmA82sxMyGVvbZzaynmb0eHvNbC3rtPjRm+wFm9rKZfWdma8zsUTNrE267DTgDOC383XjYZU61zlOxii/87B5n6htub2xmd4a/8+8tGMDuyO39O4uAEpRkt9uAuwl6OZ8BNAHeBY4H9gfuB6bGfslX4jyCIQMOBi4HrgJO3s4+VxB089IT+F/gfy0cBdWCDjSfBb4l6IB1FHAL2/97/DvBmEd54XFvAjaFx+wAvEnQF2UvgvF6WhN07ElY9lmCsXx2C6fYntUTOk8cg2OOtxvwMLCKoJsbgMfCz3gGQfdUTwIvmVm8nudFylEVn2Szu9z9mQrr7o6Zv8/MBhJ03jqriuO85+43hfNLzewigg4zn65inxfc/YFw/g4zG03Qg/l7BAmyI8FwFmsBzOwqgr7X4grv2DoAL7v7R+Hqj2OK/A54293/O2afEUCxmR3o7u+Hd5INwrvPmp6nnNjqTTM7hyARHebu682sG0Ei391/7q37LjM7BriAYHRhkUopQUk2K4hdCO9criHoYbk9wbDgjQiG0q7K+xWWV/PzQHI12ee/gBWlySk0myq4u5vZ3cCjZjYS+BfwlLsvDYv0Ag4zs+/i7L53nHhqep64zOwQgtFcz3b392Jiqgcsi6kRheB3XtkdmUgZVfFJNvu+wvI1wCXArcARBL1fzyBIVFWp2LjC2f7fTlX7GDUYVM7dxxFUTc4ADgcWxTxDq0fQg3ePClMXgtGGk3WebZjZngR3kze5+z9iNtUj+D30rBDTfsBF1YlJ6ibdQUldcijwtLs/DmBm9YCuwCdpjqMI6GRmbdx9XbiuTyI7htVuHwF3m9nDBEO4PEZQdTiIYGC5LZXs/hPB3cuOnKccC8aKeg74P3e/ucLm94AcoLW7v5PIeUVi6Q5K6pIlwLFmdkj4kP5BgiGs0+1FgkHgpprZgWbWj6BBh1PJnZWZtTCze82sv5l1NLNfAocQDBwHQUOM3YDHzay3me1lZseY2SQLBu4EWAH8wsy6mFlr+3k01Oqcp6LJBP/RvcbMdo2Zctz9A8IRms3sFAsGEe0dtnA8sfq/NqlrlKCkLrme4FlMPsEQ8WsJhgZPq/AF1yHAzgSt7h4Cbgw3/1jJbpsJnmE9QpBopxOMaHpVeMxPCYbhbkTw+RYC9wLfAaV3VPcTtM6bB6wjaKVXrfPE0R/oTpD8imOmXuH2YcDjwF0Ed2TPAX0JErRIldSThEgEmNnBBE3g93f3RZmORyQKlKBEMsDMfgV8RdCEe2/gHmCjux+c0cBEIkSNJEQyowVBa8I9gC8I3oG6PKMRiUSM7qBERCSS1EhCREQiSQlKREQiSQlKREQiSQlKREQiSQlKREQi6f8DU44G83QIunYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1fa02c99cf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "def plot_learning_curves(model, X, y):\n",
    "    X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.2, random_state=10)\n",
    "    train_errors, val_errors = [], []\n",
    "    for m in range(1, len(X_train)):\n",
    "        model.fit(X_train[:m], y_train[:m])\n",
    "        y_train_predict = model.predict(X_train[:m])\n",
    "        y_val_predict = model.predict(X_val)\n",
    "        train_errors.append(mean_squared_error(y_train_predict, y_train[:m]))\n",
    "        val_errors.append(mean_squared_error(y_val_predict, y_val))\n",
    "\n",
    "    plt.plot(np.sqrt(train_errors), \"r-+\", linewidth=2, label=\"Training set\")\n",
    "    plt.plot(np.sqrt(val_errors), \"b-\", linewidth=3, label=\"Validation set\")\n",
    "    plt.legend(loc=\"upper right\", fontsize=14)\n",
    "    plt.xlabel(\"Training set size\", fontsize=14)\n",
    "    plt.ylabel(\"RMSE\", fontsize=14)\n",
    "\n",
    "lin_reg = LinearRegression()\n",
    "plot_learning_curves(lin_reg, varData, tarData)\n",
    "plt.axis([0, 8000, 0, 20])\n",
    "save_fig(\"underfitting_learning_curves_plot\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Neural Network:**<a id = 'Neural_Network'></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As the difference between prediction to actual measurement is about two ohms, it has met our modeling goals. As an alternative approacy, let's try neural net modeling below"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_10 (Dense)             (None, 64)                704       \n",
      "_________________________________________________________________\n",
      "dropout_7 (Dropout)          (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dense_11 (Dense)             (None, 64)                4160      \n",
      "_________________________________________________________________\n",
      "dropout_8 (Dropout)          (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dense_12 (Dense)             (None, 1)                 65        \n",
      "=================================================================\n",
      "Total params: 4,929\n",
      "Trainable params: 4,929\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout\n",
    "\n",
    "numInps = len(varList)\n",
    "nnetMdl = Sequential()\n",
    "# input layer\n",
    "nnetMdl.add(Dense(units=64, activation='relu', input_dim=numInps))\n",
    "\n",
    "# hidden layers\n",
    "nnetMdl.add(Dropout(0.3, noise_shape=None, seed=None))\n",
    "nnetMdl.add(Dense(64, activation = \"relu\"))\n",
    "nnetMdl.add(Dropout(0.2, noise_shape=None, seed=None))\n",
    "          \n",
    "# output layer\n",
    "nnetMdl.add(Dense(units=1, activation='sigmoid'))\n",
    "nnetMdl.compile(loss='mean_squared_error', optimizer='adam')\n",
    "\n",
    "# Provide some info\n",
    "#from keras.utils import plot_model\n",
    "#plot_model(nnetMdl, to_file= workDir + 'model.png')\n",
    "nnetMdl.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Prepare Training (tran) and Validation (test) dataset\n",
    "varTran, varTest, tarTran, tarTest = train_test_split(varData, tarData, test_size=0.2)\n",
    "\n",
    "# scale the data\n",
    "from sklearn import preprocessing\n",
    "varScal = preprocessing.MinMaxScaler()\n",
    "varTran = varScal.fit_transform(varTran)\n",
    "varTest = varScal.transform(varTest)\n",
    "\n",
    "tarScal = preprocessing.MinMaxScaler()\n",
    "tarTran = tarScal.fit_transform(tarTran.values.reshape(-1, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 7195 samples, validate on 800 samples\n",
      "Epoch 1/50\n",
      "7195/7195 [==============================] - 0s 43us/step - loss: 0.0457 - val_loss: 0.0205\n",
      "Epoch 2/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0202 - val_loss: 0.0150\n",
      "Epoch 3/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0173 - val_loss: 0.0157\n",
      "Epoch 4/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0169 - val_loss: 0.0134\n",
      "Epoch 5/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0148 - val_loss: 0.0111\n",
      "Epoch 6/50\n",
      "7195/7195 [==============================] - 0s 6us/step - loss: 0.0129 - val_loss: 0.0094\n",
      "Epoch 7/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0110 - val_loss: 0.0078\n",
      "Epoch 8/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0093 - val_loss: 0.0061\n",
      "Epoch 9/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0078 - val_loss: 0.0045\n",
      "Epoch 10/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0063 - val_loss: 0.0034\n",
      "Epoch 11/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0055 - val_loss: 0.0026\n",
      "Epoch 12/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0047 - val_loss: 0.0021\n",
      "Epoch 13/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0043 - val_loss: 0.0018\n",
      "Epoch 14/50\n",
      "7195/7195 [==============================] - 0s 6us/step - loss: 0.0039 - val_loss: 0.0017\n",
      "Epoch 15/50\n",
      "7195/7195 [==============================] - 0s 6us/step - loss: 0.0037 - val_loss: 0.0016\n",
      "Epoch 16/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0034 - val_loss: 0.0015\n",
      "Epoch 17/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0032 - val_loss: 0.0015\n",
      "Epoch 18/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0030 - val_loss: 0.0015\n",
      "Epoch 19/50\n",
      "7195/7195 [==============================] - ETA: 0s - loss: 0.002 - 0s 8us/step - loss: 0.0029 - val_loss: 0.0015\n",
      "Epoch 20/50\n",
      "7195/7195 [==============================] - 0s 6us/step - loss: 0.0028 - val_loss: 0.0014\n",
      "Epoch 21/50\n",
      "7195/7195 [==============================] - 0s 6us/step - loss: 0.0028 - val_loss: 0.0014\n",
      "Epoch 22/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0026 - val_loss: 0.0014\n",
      "Epoch 23/50\n",
      "7195/7195 [==============================] - 0s 6us/step - loss: 0.0025 - val_loss: 0.0013\n",
      "Epoch 24/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0024 - val_loss: 0.0013\n",
      "Epoch 25/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0024 - val_loss: 0.0013\n",
      "Epoch 26/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0023 - val_loss: 0.0012\n",
      "Epoch 27/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0023 - val_loss: 0.0012\n",
      "Epoch 28/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0022 - val_loss: 0.0012\n",
      "Epoch 29/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0022 - val_loss: 0.0012\n",
      "Epoch 30/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0020 - val_loss: 0.0011\n",
      "Epoch 31/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0021 - val_loss: 0.0011\n",
      "Epoch 32/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0020 - val_loss: 0.0011\n",
      "Epoch 33/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0020 - val_loss: 0.0011\n",
      "Epoch 34/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0019 - val_loss: 0.0010\n",
      "Epoch 35/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0018 - val_loss: 0.0010\n",
      "Epoch 36/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0018 - val_loss: 9.9226e-04\n",
      "Epoch 37/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0018 - val_loss: 9.8783e-04\n",
      "Epoch 38/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0017 - val_loss: 9.5433e-04\n",
      "Epoch 39/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0017 - val_loss: 9.5077e-04\n",
      "Epoch 40/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0017 - val_loss: 9.2018e-04\n",
      "Epoch 41/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0016 - val_loss: 9.1261e-04\n",
      "Epoch 42/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0016 - val_loss: 9.0183e-04\n",
      "Epoch 43/50\n",
      "7195/7195 [==============================] - 0s 4us/step - loss: 0.0016 - val_loss: 8.8474e-04\n",
      "Epoch 44/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0016 - val_loss: 8.6719e-04\n",
      "Epoch 45/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0015 - val_loss: 8.4854e-04\n",
      "Epoch 46/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0016 - val_loss: 8.2477e-04\n",
      "Epoch 47/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0015 - val_loss: 8.0656e-04\n",
      "Epoch 48/50\n",
      "7195/7195 [==============================] - 0s 6us/step - loss: 0.0014 - val_loss: 8.1118e-04\n",
      "Epoch 49/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0014 - val_loss: 7.8844e-04\n",
      "Epoch 50/50\n",
      "7195/7195 [==============================] - 0s 5us/step - loss: 0.0014 - val_loss: 7.9169e-04\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "1.982441126153167"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hist = nnetMdl.fit(varTran, tarTran, epochs=50, batch_size=1000, validation_split=0.1)\n",
    "tarTemp = nnetMdl.predict(varTest, batch_size=1000)\n",
    "predict = tarScal.inverse_transform(tarTemp)\n",
    "resRMSE = np.sqrt(mean_squared_error(tarTest, predict))\n",
    "resRMSE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1fa042707b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(hist.history['loss'])\n",
    "plt.title('Model loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Val'], loc='upper right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With epoc increased to 100, we can even obtain 1.5 ohms accuracy. It seems this neural network model is comparable to the polynominal regressor and meet our needs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved model to disk\n"
     ]
    }
   ],
   "source": [
    "# save model and architecture to single file\n",
    "nnetMdl.save(workDir + \"LStkMdl.h5\")\n",
    "\n",
    "# finally\n",
    "print(\"Saved model to disk\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Deploy:**<a id = 'Deploy'></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the SESL3 data set as an example, we follow the similar process and built 10+ prediction models for different stakup structure setup. The polynominal model or neural network can be implemented in Java/C++ to avoid dependencies on python's package for distribution purpose. The implemented front-end, shown below, provide a quick and easy method for system designer for stackup/routing planning:\n",
    "![Deploy](https://github.com/SPISim/ML_LStkModeling/blob/master/assets/images/SLSE3Deploy.png?raw=true)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Conclusion:**<a id = 'Conclusion'></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this post/markdown document, we decribe the stackup modeling process using data science/machine learning techniques. The outcome is a deployed front-end with modeled neural network for user's instant performance evaluation. The data set and this markdown document is published on this project's git-hub page."
   ]
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}