jb-apps/Split Dataset for Model Development.ipynb

## Split Dataset for Model Development.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from random import sample\n",
    "\n",
    "from itertools import chain\n",
    "from random import sample \n",
    "import scipy\n",
    "\n",
    "import sklearn.model_selection as skl"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Read the Data\n",
    "First read in the dataframe. You'll notice it's similar to the dataframe that you ended the final solution with in Lesson 2, Exercise 4, only with more data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>Unnamed: 0</th>\n",
       "      <th>Finding Labels</th>\n",
       "      <th>Patient ID</th>\n",
       "      <th>Patient Age</th>\n",
       "      <th>Patient Gender</th>\n",
       "      <th>Atelectasis</th>\n",
       "      <th>Cardiomegaly</th>\n",
       "      <th>Consolidation</th>\n",
       "      <th>Edema</th>\n",
       "      <th>Effusion</th>\n",
       "      <th>Emphysema</th>\n",
       "      <th>Fibrosis</th>\n",
       "      <th>Hernia</th>\n",
       "      <th>Infiltration</th>\n",
       "      <th>Mass</th>\n",
       "      <th>No Finding</th>\n",
       "      <th>Nodule</th>\n",
       "      <th>Pleural_Thickening</th>\n",
       "      <th>Pneumonia</th>\n",
       "      <th>Pneumothorax</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Cardiomegaly|Emphysema</td>\n",
       "      <td>1</td>\n",
       "      <td>57</td>\n",
       "      <td>M</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>No Finding</td>\n",
       "      <td>2</td>\n",
       "      <td>77</td>\n",
       "      <td>M</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Atelectasis</td>\n",
       "      <td>3</td>\n",
       "      <td>79</td>\n",
       "      <td>M</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Cardiomegaly|Edema|Effusion</td>\n",
       "      <td>4</td>\n",
       "      <td>55</td>\n",
       "      <td>F</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Consolidation|Mass</td>\n",
       "      <td>5</td>\n",
       "      <td>68</td>\n",
       "      <td>M</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0               Finding Labels  Patient ID  Patient Age  \\\n",
       "0           1       Cardiomegaly|Emphysema           1           57   \n",
       "1           2                   No Finding           2           77   \n",
       "2           3                  Atelectasis           3           79   \n",
       "3           4  Cardiomegaly|Edema|Effusion           4           55   \n",
       "4           5           Consolidation|Mass           5           68   \n",
       "\n",
       "  Patient Gender  Atelectasis  Cardiomegaly  Consolidation  Edema  Effusion  \\\n",
       "0              M          0.0           1.0            0.0    0.0       0.0   \n",
       "1              M          0.0           0.0            0.0    0.0       0.0   \n",
       "2              M          1.0           0.0            0.0    0.0       0.0   \n",
       "3              F          0.0           1.0            0.0    1.0       1.0   \n",
       "4              M          0.0           0.0            1.0    0.0       0.0   \n",
       "\n",
       "   Emphysema  Fibrosis  Hernia  Infiltration  Mass  No Finding  Nodule  \\\n",
       "0        1.0       0.0     0.0           0.0   0.0         0.0     0.0   \n",
       "1        0.0       0.0     0.0           0.0   0.0         1.0     0.0   \n",
       "2        0.0       0.0     0.0           0.0   0.0         0.0     0.0   \n",
       "3        0.0       0.0     0.0           0.0   0.0         0.0     0.0   \n",
       "4        0.0       0.0     0.0           0.0   1.0         0.0     0.0   \n",
       "\n",
       "   Pleural_Thickening  Pneumonia  Pneumothorax  \n",
       "0                 0.0        0.0           0.0  \n",
       "1                 0.0        0.0           0.0  \n",
       "2                 0.0        0.0           0.0  \n",
       "3                 0.0        0.0           0.0  \n",
       "4                 0.0        0.0           0.0  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d = pd.read_csv('findings_data_5000.csv')\n",
    "d.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Understand the Distribution\n",
    "Just like in Lesson 2, Exercise 4, we want to see how different diseases are distributed with our disease of interest, as well as how age and gender are distributed:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_labels = np.unique(list(chain(*d['Finding Labels'].map(lambda x: x.split('|')).tolist())))\n",
    "all_labels = [x for x in all_labels if len(x)>0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0, 0.5, 'Number of Images with Label')]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = d[all_labels].sum().plot(kind='bar')\n",
    "ax.set(ylabel = 'Number of Images with Label')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Since there are many combinations of potential findings, let's look at the 30 most common co-occurrences:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f657c06ed90>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,6))\n",
    "d[d.Pneumothorax==1]['Finding Labels'].value_counts()[0:30].plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f657c01b050>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6,6))\n",
    "d[d.Pneumothorax ==1]['Patient Gender'].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([12., 18., 13., 21., 20., 47., 50., 26.,  9.,  4.]),\n",
       " array([ 7. , 14.6, 22.2, 29.8, 37.4, 45. , 52.6, 60.2, 67.8, 75.4, 83. ]),\n",
       " <a list of 10 Patch objects>)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "plt.hist(d[d.Pneumothorax==1]['Patient Age'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Split the Data into Train/Test Partitions\n",
    "Now, knowing what we know from above, let's create the appropriate training and validation sets for a model that we want to train to classify the presence of a Pneumothorax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df, valid_df = skl.train_test_split(d, \n",
    "                                   test_size = 0.2, \n",
    "                                   stratify = d['Pneumothorax'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.04401100275068767"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['Pneumothorax'].sum()/len(train_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.044"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "valid_df['Pneumothorax'].sum()/len(valid_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Great, our train_test_split made sure that we had the same proportions of Pneumothorax in both sets!\n",
    "\n",
    "### Condition 1 - To have _EQUAL_ amount of positive and negative cases of Pneumothorax in Training \n",
    "But.... we know that we want our model to be trained on a set that has _equal_ proportions of pneumothorax and no pneumothorax, so we're going to have to throw away some data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "p_inds = train_df[train_df.Pneumothorax==1].index.tolist()\n",
    "np_inds = train_df[train_df.Pneumothorax==0].index.tolist()\n",
    "\n",
    "np_sample = sample(np_inds,len(p_inds))\n",
    "train_df = train_df.loc[p_inds + np_sample]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['Pneumothorax'].sum()/len(train_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ta-da! We randomly chose a set of non-Pneumothorax images using the sample() function that was the same length as the number of true Pneumothorax cases we had, and then we threw out the rest of the non-Pneumothorax cases. Now our training dataset is balanced 50-50.\n",
    "\n",
    "### Condition 2 - To have 20% positive cases of Pneumothorax in the Test Set\n",
    "Finally, we want to make the balance in our validation set more like 20-80 since our exercise told us that the prevalence of Pneumothorax in this clinical situation is about 20%:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "p_inds = valid_df[valid_df.Pneumothorax==1].index.tolist()\n",
    "np_inds = valid_df[valid_df.Pneumothorax==0].index.tolist()\n",
    "\n",
    "# The following code pulls a random sample of non-pneumonia data that's 4 times as big as the pneumonia sample.\n",
    "np_sample = sample(np_inds,4*len(p_inds))\n",
    "valid_df = valid_df.loc[p_inds + np_sample]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "valid_df['Pneumothorax'].sum()/len(valid_df)"
   ]
  },
  {
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}