writeup.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "colnames = ['center', 'left', 'right', 'steering', 'throttle', 'brake', 'speed']\n",
    "data     = pd.read_csv('driving_log.csv', skiprows=[0], names=colnames)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Steering angle')"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20,10))\n",
    "data['steering'].plot.hist(bins=50)\n",
    "plt.ylabel('Number of samples')\n",
    "plt.xlabel('Steering angle')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Steering angle')"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJwAAAJQCAYAAADL1H4pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzs3X+w5Xdd3/HXmw0/rIJJYKExP9yg26lBx8isgQ6OYNCQQCWRwjTUSkzTrk5DxdFRE2uNihlhLKYwBWqUSHAsIUWFFVIxhl91Wkg2EEJCzGQFNGsiWZvID5HYkHf/uN+VG3J/nCSfc+89u4/HzJ17vp/zPee+d5Izd/c53x/V3QEAAACAUR612QMAAAAAcGgRnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGOmKzB5iHJz3pSb1jx47NHgMAAADgkHH99df/dXdvn2XfQzI47dixI3v37t3sMQAAAAAOGVX157Pu65Q6AAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoY7Y7AEAABbJjgvePfT9Pv2qFwx9PwCArcARTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUHMPTlW1rao+WlXvmrZPrKoPV9VtVfW2qnrMtP7YaXvf9PyOZe9x4bR+a1U9b94zAwAAAPDwbcQRTq9Icsuy7VcnuaS7dya5J8l50/p5Se7p7m9Ocsm0X6rqpCRnJ3laktOTvKGqtm3A3AAAAAA8DHMNTlV1XJIXJPnNabuSnJrk7dMulyc5a3p85rSd6fnnTvufmeSK7r63uz+VZF+SU+Y5NwAAAAAP37yPcPovSX46yf3T9hOT/E133zdt709y7PT42CS3J8n0/Gen/f9hfYXX/IOq2l1Ve6tq74EDB0b/OQAAAACY0dyCU1X98yR3dff1y5dX2LXXeW6t13xlofvS7t7V3bu2b9/+kOcFAAAAYIwj5vjez0rywqp6fpLHJXlClo54OrKqjpiOYjouyR3T/vuTHJ9kf1UdkeTrk9y9bP2g5a8BAAAAYIuZ2xFO3X1hdx/X3TuydNHv93b3DyZ5X5IXT7udk+Sd0+M903am59/b3T2tnz3dxe7EJDuTXDuvuQEAAAB4ZOZ5hNNqfibJFVX1y0k+muRN0/qbkvx2Ve3L0pFNZydJd99cVVcm+USS+5Kc391f3vixAQAAAJjFhgSn7n5/kvdPjz+ZFe4y191fSvKSVV5/cZKL5zchAAAAAKPM+y51AAAAABxmBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKHmFpyq6nFVdW1Vfayqbq6qX5zW31xVn6qqG6avk6f1qqrXVdW+qrqxqp6+7L3Oqarbpq9z5jUzAAAAAI/cEXN873uTnNrdX6iqRyf5k6r6n9NzP9Xdb/+q/c9IsnP6ekaSNyZ5RlUdneSiJLuSdJLrq2pPd98zx9kBAAAAeJjmdoRTL/nCtPno6avXeMmZSd4yve5DSY6sqmOSPC/J1d199xSZrk5y+rzmBgAAAOCRmes1nKpqW1XdkOSuLEWjD09PXTydNndJVT12Wjs2ye3LXr5/WlttHQAAAIAtaK7Bqbu/3N0nJzkuySlV9a1JLkzyT5N8Z5Kjk/zMtHut9BZrrD9AVe2uqr1VtffAgQND5gcAAADgoduQu9R1998keX+S07v7zum0uXuT/FaSU6bd9ic5ftnLjktyxxrrX/0zLu3uXd29a/v27XP4UwAAAAAwi3nepW57VR05Pf6aJN+b5E+n6zKlqirJWUluml6yJ8nLprvVPTPJZ7v7ziTvSXJaVR1VVUclOW1aAwAAAGALmudd6o5JcnlVbctS2Lqyu99VVe+tqu1ZOlXuhiQ/Ou1/VZLnJ9mX5ItJzk2S7r67ql6Z5Lppv1/q7rvnODcAAAAAj8DcglN335jkO1ZYP3WV/TvJ+as8d1mSy4YOCAAAAMBcbMg1nAAAAAA4fAhOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUIITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUIITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUIITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUIITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADDW34FRVj6uqa6vqY1V1c1X94rR+YlV9uKpuq6q3VdVjpvXHTtv7pud3LHuvC6f1W6vqefOaGQAAAIBHbp5HON2b5NTu/vYkJyc5vaqemeTVSS7p7p1J7kly3rT/eUnu6e5vTnLJtF+q6qQkZyd5WpLTk7yhqrbNcW4AAAAAHoG5Bade8oVp89HTVyc5Ncnbp/XLk5w1PT5z2s70/HOrqqb1K7r73u7+VJJ9SU6Z19wAAAAAPDJzvYZTVW2rqhuS3JXk6iR/luRvuvu+aZf9SY6dHh+b5PYkmZ7/bJInLl9f4TUAAAAAbDFzDU7d/eXuPjnJcVk6KulbVtpt+l6rPLfa+gNU1e6q2ltVew8cOPBwRwYAAADgEdqQu9R1998keX+SZyY5sqqOmJ46Lskd0+P9SY5Pkun5r09y9/L1FV6z/Gdc2t27unvX9u3b5/HHAAAAAGAG87xL3faqOnJ6/DVJvjfJLUnel+TF027nJHnn9HjPtJ3p+fd2d0/rZ093sTsxyc4k185rbgAAAAAemSPW3+VhOybJ5dMd5R6V5MrufldVfSLJFVX1y0k+muRN0/5vSvLbVbUvS0c2nZ0k3X1zVV2Z5BNJ7ktyfnd/eY5zAwAAAPAIzC04dfeNSb5jhfVPZoW7zHX3l5K8ZJX3ujjJxaNnBAAAAGC8DbmGEwAAAACHD8EJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGCodYNTVb2iqp5QS95UVR+pqtM2YjgAAAAAFs8sRzj9m+7+XJLTkmxPcm6SV633oqo6vqreV1W3VNXNVfWKaf0Xquovq+qG6ev5y15zYVXtq6pbq+p5y9ZPn9b2VdUFD/lPCQAAAMCGOWKGfWr6/vwkv9XdH6uqWusFk/uS/GR3f6SqHp/k+qq6enruku7+zw/4IVUnJTk7ydOSfEOSP66qfzI9/fok35dkf5LrqmpPd39ihhkAAAAA2GCzBKfrq+qPkpyY5MIpHt2/3ou6+84kd06PP19VtyQ5do2XnJnkiu6+N8mnqmpfklOm5/Z19yeTpKqumPYVnAAAAAC2oFlOqTsvyQVJvrO7v5jkMVk6rW5mVbUjyXck+fC09PKqurGqLquqo6a1Y5Pcvuxl+6e11dYBAAAA2IJmCU6d5KQkPzZtf22Sx836A6rq65L8bpIfn64F9cYk35Tk5CwdAfWag7uu8rNXW//qn7O7qvZW1d4DBw7MOh4AAAAAg80SnN6Q5J8leem0/fksXVNpXVX16CzFpt/p7t9Lku7+THd/ubvvT/Ib+cppc/uTHL/s5ccluWON9Qfo7ku7e1d379q+ffss4wEAAAAwB7MEp2d09/lJvpQk3X1Plk6rW9N0YfE3Jbmlu39t2foxy3b7gSQ3TY/3JDm7qh5bVScm2Znk2iTXJdlZVSdW1WOydGHxPTPMDQAAAMAmmOWi4f+vqrZlOo2tqrZnhouGJ3lWkh9K8vGqumFa+9kkL62qk6f3+3SSH0mS7r65qq7M0sXA70tyfnd/efqZL0/yniTbklzW3TfP9scDAAAAYKPNEpxel+T3kzy5qi5O8uIkP7fei7r7T7Ly9ZeuWuM1Fye5eIX1q9Z6HQAAAABbx7rBqbt/p6quT/LcLAWks7r7lrlPBgAAAMBCWjU4VdXRyzbvSvLW5c91993zHAwAAACAxbTWEU7XZ+k6SyudFtdJnjqXiQAAAABYaKsGp+4+cSMHAQAAAODQMMtFw1NVL0ryXVk6sul/dfc75joVAAAAAAvrUevtUFVvSPKjST6e5KYkP1pVr5/3YAAAAAAsplmOcHp2km/t7k6Sqro8S/EJAAAAAB5k3SOcktya5IRl28cnuXE+4wAAAACw6GY5wumJSW6pqmun7e9M8n+qak+SdPcL5zUcAAAAAItnluD083OfAgAAAIBDxrrBqbs/kCRV9YTl+3f33XOcCwAAAIAFtW5wqqrdSV6Z5O+S3J+kknSSp853NAAAAAAW0Syn1P1Ukqd191/PexgAAAAAFt8sd6n7syRfnPcgAAAAABwaZjnC6cIk/7uqPpzk3oOL3f1jc5sKAAAAgIU1S3D69STvTfLxLF3DCQAAAABWNUtwuq+7f2LukwAAAABwSJjlGk7vq6rdVXVMVR198GvukwEAAACwkGY5wulfTd8vXLbWSZ46fhwAAAAAFt26wam7T9yIQQAAAAA4NMxyhFOq6luTnJTkcQfXuvst8xoKAAAAgMW1bnCqqouSPCdLwemqJGck+ZMkghMAAAAADzLLRcNfnOS5Sf6qu89N8u1JHjvXqQAAAABYWLMEp7/r7vuT3FdVT0hyV1wwHAAAAIBVzHINp71VdWSS30hyfZIvJLl2rlMBAAAAsLBmuUvdv58e/req+sMkT+juG+c7FgAAAACLat1T6qrqWVX1tdPmdyX54ar6xvmOBQAAAMCimuUaTm9M8sWq+vYkP53kz+MOdQAAAACsYpbgdF93d5Izk7y2u1+b5PHzHQsAAACARTXLRcM/X1UXJvnXSb67qrYlefR8xwIAAABgUc1yhNO/THJvkvO6+6+SHJvkV+c6FQAAAAALa5a71P1Vkl9btv0XcQ0nAAAAAFYxyxFOAAAAADAzwQkAAACAoVYNTlV1zfT91Rs3DgAAAACLbq1rOB1TVc9O8sKquiJJLX+yuz8y18kAAAAAWEhrBaefT3JBkuOy7KLhk05y6ryGAgAAAGBxrRqcuvvtSd5eVf+pu1+5gTMBAAAAsMDWOsIpSdLdr6yqFyb57mnp/d39rvmOBQAAAMCiWvcudVX1K0lekeQT09crpjUAAAAAeJB1j3BK8oIkJ3f3/UlSVZcn+WiSC+c5GAAAAACLad0jnCZHLnv89fMYBAAAAIBDwyxHOP1Kko9W1fuSVJau5eToJgAAAABWNMtFw99aVe9P8p1ZCk4/091/Ne/BAAAAAFhMsxzhlO6+M8meOc8CAAAAwCFg1ms4AQAAAMBMBCcAAAAAhlozOFXVo6rqpo0aBgAAAIDFt2Zw6u77k3ysqk7YoHkAAAAAWHCzXDT8mCQ3V9W1Sf724GJ3v3BuUwEAAACwsGYJTr849ykAAAAAOGSsG5y6+wNV9Y1Jdnb3H1fVP0qybf6jAQAAALCI1r1LXVX9uyRvT/Lr09KxSd4xz6EAAAAAWFzrBqck5yd5VpLPJUl335bkyfMcCgAAAIDFNUtwure7//7gRlUdkaTXe1FVHV9V76uqW6rq5qp6xbR+dFVdXVW3Td+Pmtarql5XVfuq6saqevqy9zpn2v+2qjrnof8xAQAAANgoswSnD1TVzyb5mqr6viT/I8kfzPC6+5L8ZHd/S5JnJjm/qk5KckGSa7p7Z5Jrpu0kOSPJzulrd5I3JkuBKslFSZ6R5JQkFx2MVAAAAABsPbMEpwuSHEjy8SQ/kuSqJD+33ou6+87u/sj0+PNJbsnS9Z/OTHL5tNvlSc6aHp+Z5C295ENJjqyqY5I8L8nV3X13d9+T5Ookp8/45wMAAABgg81yl7r7q+ryJB/O0ql0t3b3uqfULVdVO5J8x/QeT+nuO6f3vrOqDl4P6tgkty972f5pbbX1r/4Zu7N0ZFROOOGEhzIeAAAAAAPNcpe6FyT5sySvS/Jfk+yrqjNm/QFV9XVJfjfJj3f359badYW1XmP9gQvdl3b3ru7etX379lnHAwAAAGCwWU6pe02S7+nu53T3s5N8T5JLZnnzqnp0lmLT73T3703Ln5lOlcv0/a5pfX+S45e9/Lgkd6yxDgAAAMAWNEtwuqu79y3b/mS+EolWVVWV5E1JbunuX1v21J4kB+80d06Sdy5bf9l0t7pnJvnsdOrde5KcVlVHTRcLP21aAwAAAGALWvUaTlX1ounhzVV1VZIrs3Qq20uSXDfDez8ryQ8l+XhV3TCt/WySVyW5sqrOS/IX0/slSxcjf36SfUm+mOTcJOnuu6vqlct+5i91992z/fEAAAAA2GhrXTT8+5c9/kySZ0+PDyQ5ar037u4/ycrXX0qS566wfyc5f5X3uizJZev9TAAAAAA236rBqbvP3chBAAAAADg0rHWEU5Kkqk5M8h+S7Fi+f3e/cH5jAQAAALCo1g1OSd6RpYt//0GS++c7DgAAAACLbpbg9KXuft3cJwEAAADgkDBLcHptVV2U5I+S3Htwsbs/MrepAAAAAFhYswSnb0vyQ0lOzVdOqetpGwAAAAAeYJbg9ANJntrdfz/vYQAAAABYfI+aYZ+PJTly3oMAAAAAcGiY5QinpyT506q6Lg+8htML5zYVAAAAAAtrluB00dynAAAAAOCQsW5w6u4PbMQgAAAAABwa1g1OVfX5LN2VLkkek+TRSf62u58wz8EAAAAAWEyzHOH0+OXbVXVWklPmNhEAAAAAC22Wu9Q9QHe/I8mpc5gFAAAAgEPALKfUvWjZ5qOS7MpXTrEDAAAAgAeY5S5137/s8X1JPp3kzLlMAwAAAMDCm+UaTuduxCAAAAAAHBpWDU5V9fNrvK67+5VzmAcAAACABbfWEU5/u8La1yY5L8kTkwhOAAAAADzIqsGpu19z8HFVPT7JK5Kcm+SKJK9Z7XUAAAAAHN7WvIZTVR2d5CeS/GCSy5M8vbvv2YjBAAAAAFhMa13D6VeTvCjJpUm+rbu/sGFTAQAAALCwHrXGcz+Z5BuS/FySO6rqc9PX56vqcxszHgAAAACLZq1rOK0VowAAAABgRaISAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUIITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUIITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAw1t+BUVZdV1V1VddOytV+oqr+sqhumr+cve+7CqtpXVbdW1fOWrZ8+re2rqgvmNS8AAAAAY8zzCKc3Jzl9hfVLuvvk6euqJKmqk5KcneRp02veUFXbqmpbktcnOSPJSUleOu0LAAAAwBZ1xLzeuLs/WFU7Ztz9zCRXdPe9ST5VVfuSnDI9t6+7P5kkVXXFtO8nBo8LAAAAwCCbcQ2nl1fVjdMpd0dNa8cmuX3ZPvuntdXWH6SqdlfV3qrae+DAgXnMDQAAAMAMNjo4vTHJNyU5OcmdSV4zrdcK+/Ya6w9e7L60u3d1967t27ePmBUAAACAh2Fup9StpLs/c/BxVf1GkndNm/uTHL9s1+OS3DE9Xm0dAAAAgC1oQ49wqqpjlm3+QJKDd7Dbk+TsqnpsVZ2YZGeSa5Ncl2RnVZ1YVY/J0oXF92zkzAAAAAA8NHM7wqmq3prkOUmeVFX7k1yU5DlVdXKWTov7dJIfSZLuvrmqrszSxcDvS3J+d395ep+XJ3lPkm1JLuvum+c1MwAAAACP3DzvUvfSFZbftMb+Fye5eIX1q5JcNXA0AAAAAOZoM+5SBwAAAMAhTHACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhhKcAAAAABhKcAIAAABgqLkFp6q6rKruqqqblq0dXVVXV9Vt0/ejpvWqqtdV1b6qurGqnr7sNedM+99WVefMa14AAAAAxpjnEU5vTnL6V61dkOSa7t6Z5JppO0nOSLJz+tqd5I3JUqBKclGSZyQ5JclFByMVAAAAAFvT3IJTd38wyd1ftXxmksunx5cnOWvZ+lt6yYeSHFlVxyR5XpKru/vu7r4nydV5cMQCAAAAYAvZ6Gs4PaW770yS6fuTp/Vjk9y+bL/909pq6w9SVburam9V7T1w4MDwwQEAAACYzVa5aHitsNZrrD94sfvS7t7V3bu2b98+dDgAAAAAZrfRwekz06lymb7fNa3vT3L8sv2OS3LHGusAAAAAbFEbHZz2JDl4p7lzkrxz2frLprvVPTPJZ6dT7t6T5LSqOmq6WPhp0xoAAAAAW9QR83rjqnprkuckeVJV7c/S3eZeleTKqjovyV8kecm0+1VJnp9kX5IvJjk3Sbr77qp6ZZLrpv1+qbu/+kLkAAAAAGwhcwtO3f3SVZ567gr7dpLzV3mfy5JcNnA0AAAAAOZoq1w0HAAAAIBDhOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUIITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUIITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUIITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDCU4AAAAADCU4AQAAADCU4AQAAADAUIITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFBHbPYAAABsbTsuePfQ9/v0q14w9P0AgK3HEU4AAAAADOUIJwAAwJFsAAy1KUc4VdWnq+rjVXVDVe2d1o6uqqur6rbp+1HTelXV66pqX1XdWFVP34yZAQAAAJjNZp5S9z3dfXJ375q2L0hyTXfvTHLNtJ0kZyTZOX3tTvLGDZ8UAAAAgJltpWs4nZnk8unx5UnOWrb+ll7yoSRHVtUxmzEgAAAAAOvbrODUSf6oqq6vqt3T2lO6+84kmb4kPDvtAAALtUlEQVQ/eVo/Nsnty167f1oDAAAAYAvarIuGP6u776iqJye5uqr+dI19a4W1ftBOS+Fqd5KccMIJY6YEAAAA4CHblCOcuvuO6ftdSX4/ySlJPnPwVLnp+13T7vuTHL/s5ccluWOF97y0u3d1967t27fPc3wAAAAA1rDhwamqvraqHn/wcZLTktyUZE+Sc6bdzknyzunxniQvm+5W98wknz146h0AAAAAW89mnFL3lCS/X1UHf/5/7+4/rKrrklxZVecl+YskL5n2vyrJ85PsS/LFJOdu/MgAAAAAzGrDg1N3fzLJt6+w/n+TPHeF9U5y/gaMBgAAAMAAm3WXOgAAAAAOUYITAAAAAEMJTgAAAAAMJTgBAAAAMJTgBAAAAMBQG36XOgCAjbTjgndv9ggAAIcdRzgBAAAAMJTgBAAAAMBQghMAAAAAQwlOAAAAAAwlOAEAAAAwlOAEAAAAwFCCEwAAAABDHbHZAwAAHM52XPDu4e/56Ve9YPh7AgA8FI5wAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYykXDAYAtZR4X0QYAYGMJTgDAwyYOAQCwEqfUAQAAADCU4AQAAADAUIITAAAAAEO5hhMAwCHGtbUAgM3mCCcAAAAAhhKcAAAAABhKcAIAAABgKMEJAAAAgKEEJwAAAACGEpwAAAAAGEpwAgAAAGAowQkAAACAoQQnAAAAAIYSnAAAAAAYSnACAAAAYCjBCQAAAIChBCcAAAAAhjpiswcAYGvYccG7h7/np1/1guHvCQAAbH2CE8CCmkcgGm30jAIWAAAsBqfUAQAAADCU4AQAAADAUIITAAAAAEO5hhMADOLC6wAAsERwAoAtzIXXAZaI+gCLxSl1AAAAAAwlOAEAAAAwlFPqAOAwMo9TUuChcmoUABz6BCcAABbeVo+pghgAhxvBCQAA5uxwvAHAVo+A83A4/ncGWI3gBAAAHJYOxygGsFFcNBwAAACAoQQnAAAAAIZySh3ABnHY/iPn2hgAALAYBCfgkCDmAAAAbB0LE5yq6vQkr02yLclvdverNnkkAACAheJoYWCjLERwqqptSV6f5PuS7E9yXVXt6e5PbO5kwMPliCS2Av8fAgDAfCxEcEpySpJ93f3JJKmqK5KcmURwgg3iH+YAsHX4vXx48N8ZWGSLEpyOTXL7su39SZ6xSbNsKIe8bk1b/Ze//84AAGyErf734nkY/Xdt/+bjULUowalWWOsH7FC1O8nuafMLVXXr3KdaQPXqzZ5gS3lSkr/e7CHmwX9n5uCQ/bzAnPjMwOx8Xlgom/x37XU/L/4twJx946w7Lkpw2p/k+GXbxyW5Y/kO3X1pkks3cigWW1Xt7e5dmz0HLAKfF3hofGZgdj4vMDufFxbJozZ7gBldl2RnVZ1YVY9JcnaSPZs8EwAAAAArWIgjnLr7vqp6eZL3JNmW5LLuvnmTxwIAAABgBQsRnJKku69KctVmz8EhxSmYMDufF3hofGZgdj4vMDufFxZGdff6ewEAAADAjBblGk4AAAAALAjBicNGVb2kqm6uqvuratU7O1TV6VV1a1Xtq6oLNnJG2Cqq6uiqurqqbpu+H7XKfl+uqhumLzdz4LCy3u+LqnpsVb1tev7DVbVj46eErWOGz8wPV9WBZb9X/u1mzAlbQVVdVlV3VdVNqzxfVfW66fN0Y1U9faNnhPUIThxObkryoiQfXG2HqtqW5PVJzkhyUpKXVtVJGzMebCkXJLmmu3cmuWbaXsnfdffJ09cLN2482Fwz/r44L8k93f3NSS5J8uqNnRK2jofwd6y3Lfu98psbOiRsLW9Ocvoaz5+RZOf0tTvJGzdgJnhIBCcOG919S3ffus5upyTZ192f7O6/T3JFkjPnPx1sOWcmuXx6fHmSszZxFtiKZvl9sfxz9PYkz62q2sAZYSvxdyx4CLr7g0nuXmOXM5O8pZd8KMmRVXXMxkwHsxGc4IGOTXL7su390xocbp7S3XcmyfT9yavs97iq2ltVH6oqUYrDySy/L/5hn+6+L8lnkzxxQ6aDrWfWv2P9i+n0oLdX1fEbMxosJP9uYcs7YrMHgJGq6o+T/OMVnvqP3f3OWd5ihTW3cuSQtNbn5SG8zQndfUdVPTXJe6vq4939Z2MmhC1tlt8XfqfAV8zyefiDJG/t7nur6kezdITgqXOfjP/f3v2H+lXXcRx/vtwP58WYySZMkgrbRNtkuVXKEh2JgYQg3Mow3KKQqTElQopGSqLYH0W5GLmVKP0xZtAfE40ptQmK1lyN/QrHkKljw7EoqaTV1rs/7md4nbt3N+9x3+u+z8c/59zP+Zz3eX8vHM73+z6f8zl6f/IaownPgpNOK1V1zThD7AOG3037ELB/nDGlCWm08yXJ60lmVdWBNjz74Agx9rfly0k2AZ8ALDipH4zlenGsz74kk4HpjP54hHQ6O+k5U1V/GfbnGpz3TBqNv1s04flInfR2m4HZST6aZCpwI+Cbt9SP1gNL2voS4B0jBJN8MMmZbX0GsAjYdcoylHprLNeL4efRIPC7qvLus/rVSc+Z4+afuR748ynMT3q/WQ/c3N5WdznwxrHpEKSJwhFO6htJbgBWAjOBJ5JsrarPJTkf+HlVXVdVR5J8A9gATAIerqqdPUxb6pUHgMeSfA14FfgCQJKFwLKq+jpwMfBQkv8ydAPjgaqy4KS+MNL1Isn3gReraj3wC+CXSfYwNLLpxt5lLPXWGM+Z5UmuB44wdM4s7VnCUo8lWQtcDcxIsg+4G5gCUFU/A54ErgP2AG8CX+1NptLI4o02SZIkSZIkdclH6iRJkiRJktQpC06SJEmSJEnqlAUnSZIkSZIkdcqCkyRJkiRJkjplwUmSJEmSJEmdsuAkSZL6TpLvJtmZZFuSrUk+3drvTDLQ4XGWJbm5q3hdSbI3yYxe5yFJkk5fqape5yBJknTKJLkC+BFwdVUdboWXqVW1P8leYGFVHergOJOr6sh447wXuvyckiRJJ+IIJ0mS1G9mAYeq6jBAVR1qxablwPnAxiQbAZJcm+T5JH9M8qskZ7f2BUmeSbIlyYYks1r7piT3J3kGuCPJPUm+NWzbD5L8IcnuJFe29oEkj7XRVuuS/D7JwuOTTvK9JJuT7EiyOkk6ivuVtu/WJA8lmdT9v1ySJPUbC06SJKnfPAVc0Iozq5JcBVBVDwL7gcVVtbiNfFoBXFNVlwEvAt9MMgVYCQxW1QLgYeC+YfHPqaqrquqHJzj25Kr6FHAncHdruw34a1VdCtwLLBgh759W1Serai5wFvD58cZNcjHwJWBRVc0HjgI3jXB8SZKkMZvc6wQkSZJOpar6R5IFwJXAYmBdkm9X1SPHdb0cuAR4rg0mmgo8D1wEzAWebu2TgAPD9ls3yuF/3ZZbgI+09c8AP2m57UiybYR9Fye5CxgAzgV2Ao+PM+5nGSpEbW6f5Szg4Cj5S5IkjYkFJ0mS1Heq6iiwCdiUZDuwBHjkuG4Bnq6qL7+tMZkH7KyqK0YI/89RDn24LY/y1vewnCzfJNOAVQzNu/RaknuAaeON2/o8WlXfGUNfSZKkMfOROkmS1FeSXJRk9rCm+cArbf3vwAfa+gvAoiQfa/sNJJkDvATMbJOPk2RKko+PI6VngS+2WJcA807Q51hx6VCbR2qwo7i/BQaTnNf6nZvkw/9f+pIkSe/kCCdJktRvzgZWJjkHOALsAW5p21YDv0lyoM3jtBRYm+TMtn1FVe1OMgg8mGQ6Q9+nfszQI27vxirg0fbI25+AbcAbwztU1d+SrAG2A3uBzR3F3ZVkBfBUkjOA/wC381YBTpIk6V1JVfU6B0mSpL7V3go3par+leRChkYdzamqf0/EuJIkSWPhCCdJkqTeGgA2trffBbi1o6LQexVXkiTppBzhJEmSJEmSpE45abgkSZIkSZI6ZcFJkiRJkiRJnbLgJEmSJEmSpE5ZcJIkSZIkSVKnLDhJkiRJkiSpUxacJEmSJEmS1Kn/AQkD+fZS6644AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import cv2\n",
    "from sklearn.utils import shuffle\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "\n",
    "# Constants\n",
    "steering_adjustment = 0.27\n",
    "\n",
    "# What determines a turn, Can be done better by taking a mean or some other metric\n",
    "# of the positive values and in turn for the negative values\n",
    "turn_thresh         = 0.15\n",
    "turn_angle_noise    = (-0.10, 0.10)\n",
    "\n",
    "# Preparing data from driving_log.csv (Sample training data)\n",
    "colnames = ['center', 'left', 'right', 'steering', 'throttle', 'brake', 'speed']\n",
    "data     = pd.read_csv('driving_log.csv', skiprows=[0], names=colnames)\n",
    "\n",
    "center = np.array([x.strip() for x in data.center.tolist()])\n",
    "left = np.array([x.strip() for x in data.left.tolist()])\n",
    "right = np.array([x.strip() for x in data.right.tolist()])\n",
    "steering = np.array(data.steering.tolist())\n",
    "\n",
    "# input_shape = mpimg.imread(center[0]).shape\n",
    "\n",
    "# Split train and valid\n",
    "# Shuffle center and steering. Using 10% of center camera images and steering angles for validation.\n",
    "center_train, X_valid, steering_train, y_valid = train_test_split(center, steering, test_size=0.10, shuffle=True, random_state=100)\n",
    "\n",
    "center_train = np.array(center_train)\n",
    "steering_train = np.array(steering_train)\n",
    "\n",
    "# Filtering going straight, left and right turns\n",
    "a_right = steering_train[steering_train > turn_thresh]\n",
    "d_right = center_train[steering_train > turn_thresh]\n",
    "\n",
    "a_left = steering_train[steering_train < -turn_thresh]\n",
    "d_left = center_train[steering_train < -turn_thresh]\n",
    "\n",
    "a_straight = steering_train[(steering_train < turn_thresh ) & (steering_train > -turn_thresh)]\n",
    "d_straight = center_train[(steering_train < turn_thresh ) & (steering_train > -turn_thresh)]\n",
    "\n",
    "# To account for the discrepancy in the dataset between the straight, right, left turns\n",
    "# Since the number of images with steering straight is much higher than the rest,\n",
    "# turns aren't able to contribute to the CNN weights effectively\n",
    "\n",
    "# Filter angles less than -turn_thresh and add right camera images into driving left list, with a sharper turn angle\n",
    "# Left camera images can be added to the left driving list by making a softer turn angle than the center image\n",
    "# The opposite is done for driving right list\n",
    "# Could be done better than converting to list but dont have to worry about appending dimensions\n",
    "a_left = a_left.tolist() + (steering[np.where(steering < -turn_thresh)] - steering_adjustment).tolist()\n",
    "d_left = d_left.tolist() + (right[np.where(steering < -turn_thresh)]).tolist()\n",
    "\n",
    "a_right = a_right.tolist() + (steering[np.where(steering > turn_thresh)] + steering_adjustment).tolist()\n",
    "d_right = d_right.tolist() + (left[np.where(steering > turn_thresh)]).tolist()\n",
    "\n",
    "# Combine images from center, left and right cameras\n",
    "X_train = d_straight.tolist() + d_left + d_right\n",
    "y_train = np.float32(a_straight.tolist() + a_left + a_right)\n",
    "\n",
    "\n",
    "plt.figure(figsize=(20,10))\n",
    "plt.hist(y_train, bins=50)\n",
    "plt.ylabel('Number of samples')\n",
    "plt.xlabel('Steering angle')\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pickle as p\n",
    "\n",
    "hist = p.load(open('hist.p','rb'))\n",
    "\n",
    "### plot the training and validation loss for each epoch\n",
    "plt.figure(figsize=(20,10))\n",
    "plt.plot(hist['loss'])\n",
    "plt.plot(hist['val_loss'])\n",
    "plt.title('model mean squared error loss')\n",
    "plt.ylabel('mean squared error loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['training set', 'validation set'], loc='upper right')\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}