{ "cells": [ { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "2C0MKJaprzeU" }, "source": [ "[![AnalyticsDojo](https://github.com/rpi-techfundamentals/fall2018-materials/blob/master/fig/final-logo.png?raw=1)](http://rpi.analyticsdojo.com)\n", "

Fastai - Revisiting Titanic

\n", "

rpi.analyticsdojo.com

\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Titanic Fastai" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "jo_XONekrzep" }, "outputs": [], "source": [ "from fastai import *\n", "from fastai.tabular import * \n", "import numpy as np\n", "import pandas as pd\n", "import pandas as pd\n", "\n", "train= pd.read_csv('https://raw.githubusercontent.com/rpi-techfundamentals/fall2018-materials/master/input/train.csv')\n", "test = pd.read_csv('https://raw.githubusercontent.com/rpi-techfundamentals/fall2018-materials/master/input/test.csv')\n", "\n" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 302 }, "colab_type": "code", "id": "FfTX5TLgVOLb", "outputId": "319c0f98-29f2-4862-8b6d-7da06f3b43c1" }, "outputs": [ { "data": { "text/plain": [ "PassengerId 0\n", "Survived 0\n", "Pclass 0\n", "Name 0\n", "Sex 0\n", "Age 0\n", "SibSp 0\n", "Parch 0\n", "Ticket 0\n", "Fare 0\n", "Cabin 687\n", "Embarked 0\n", "Title 0\n", "NameLength 0\n", "FamilyS 0\n", "dtype: int64" ] }, "execution_count": 2, "metadata": { "tags": [] }, "output_type": "execute_result" } ], "source": [ "#Create a categorical variable from the family count \n", "def family(x):\n", " if x < 2:\n", " return 'Single'\n", " elif x == 2:\n", " return 'Couple'\n", " elif x <= 4:\n", " return 'InterM'\n", " else:\n", " return 'Large'\n", "\n", "\n", "for df in [train, test]:\n", " df['Title'] = df['Name'].str.split(',').str[1].str.split(' ').str[1]\n", " df['Title'] = df['Title'].replace(['Lady', 'the Countess', 'Countess', 'Capt', 'Col', 'Don', 'Dr', 'Major', 'Rev', 'Sir', 'Jonkheer', 'Dona', 'Ms', 'Mme', 'Mlle'], 'Rare')\n", " df['Age']=df['Age'].fillna(df['Age'].median())\n", " df['Fare']=df['Fare'].fillna(df['Fare'].median())\n", " df['Embarked']=df['Embarked'].fillna('S')\n", " df['NameLength'] = df['Name'].map(lambda x: len(x))\n", " df['FamilyS'] = df['SibSp'] + df['Parch'] + 1\n", " df['FamilyS'] = df['FamilyS'].apply(family)\n", "train.isnull().sum(axis = 0)\n" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 493 }, "colab_type": "code", "id": "gSR2qKc4YPkT", "outputId": "2722bd27-9459-4fb6-ed61-e035dc2728e2" }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PassengerIdSurvivedPclassNameSexAgeSibSpParchTicketFareCabinEmbarkedTitleNameLengthFamilyS
0103Braund, Mr. Owen Harrismale22.010A/5 211717.2500NaNSMr.23Couple
1211Cumings, Mrs. John Bradley (Florence Briggs Th...female38.010PC 1759971.2833C85CMrs.51Couple
2313Heikkinen, Miss. Lainafemale26.000STON/O2. 31012827.9250NaNSMiss.22Single
3411Futrelle, Mrs. Jacques Heath (Lily May Peel)female35.01011380353.1000C123SMrs.44Couple
4503Allen, Mr. William Henrymale35.0003734508.0500NaNSMr.24Single
\n", "
" ], "text/plain": [ " PassengerId Survived Pclass \\\n", "0 1 0 3 \n", "1 2 1 1 \n", "2 3 1 3 \n", "3 4 1 1 \n", "4 5 0 3 \n", "\n", " Name Sex Age SibSp \\\n", "0 Braund, Mr. Owen Harris male 22.0 1 \n", "1 Cumings, Mrs. John Bradley (Florence Briggs Th... female 38.0 1 \n", "2 Heikkinen, Miss. Laina female 26.0 0 \n", "3 Futrelle, Mrs. Jacques Heath (Lily May Peel) female 35.0 1 \n", "4 Allen, Mr. William Henry male 35.0 0 \n", "\n", " Parch Ticket Fare Cabin Embarked Title NameLength FamilyS \n", "0 0 A/5 21171 7.2500 NaN S Mr. 23 Couple \n", "1 0 PC 17599 71.2833 C85 C Mrs. 51 Couple \n", "2 0 STON/O2. 3101282 7.9250 NaN S Miss. 22 Single \n", "3 0 113803 53.1000 C123 S Mrs. 44 Couple \n", "4 0 373450 8.0500 NaN S Mr. 24 Single " ] }, "execution_count": 3, "metadata": { "tags": [] }, "output_type": "execute_result" } ], "source": [ "train.head()" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "PKz107Q3XsSL" }, "outputs": [], "source": [ "dep_var = 'Survived'\n", "cat_names = ['Pclass', 'Sex', 'Embarked', 'Title', 'FamilyS']\n", "cont_names = ['Age', 'Fare', 'SibSp', 'Parch', 'NameLength']\n", "procs = [FillMissing, Categorify, Normalize]\n", "test_data = (TabularList.from_df(test, path='.', cat_names=cat_names, cont_names=cont_names, procs=procs))\n", "\n" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "r2P_Y1Ipd_k5" }, "outputs": [], "source": [ "\n", "data = (TabularList.from_df(train, path='.', cat_names=cat_names, cont_names=cont_names, procs=procs)\n", " .split_by_idx(list(range(0,200)))\n", " .label_from_df(cols=dep_var)\n", " .add_test(test_data, label=0)\n", " .databunch())" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 241 }, "colab_type": "code", "id": "En78trmC5nPS", "outputId": "f7defd77-d62c-40d0-9980-45bc1eb3f442" }, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PclassSexEmbarkedTitleFamilySAgeFareSibSpParchNameLengthtarget
2maleSMr.Couple0.5653-0.14320.5043-0.4658-0.41630
3maleSMr.Single-0.3706-0.4860-0.4610-0.4658-0.63190
3maleSMr.Single-0.1366-0.4982-0.4610-0.4658-0.52410
3maleSMr.Single3.4510-0.4883-0.4610-0.4658-0.84750
3maleSMr.Couple-0.9165-0.51250.5043-0.4658-0.20080
" ], "text/plain": [ "" ] }, "metadata": { "tags": [] }, "output_type": "display_data" } ], "source": [ "#Shows the Data\n", "data.show_batch()" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "TI-X15PcStFy" }, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "eYUUnFGE54qO" }, "outputs": [], "source": [ "#Define our Learner\n", "learn = tabular_learner(data, layers=[300,100], metrics=accuracy)\n" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 449 }, "colab_type": "code", "id": "EZ0pQLyfTAqX", "outputId": "a8f9ce9b-8461-4131-d6ee-0a02a73cfc6a" }, "outputs": [ { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": { "tags": [] }, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "LR Finder is complete, type {learner_name}.recorder.plot() to see the graph.\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZQAAAEKCAYAAAA1qaOTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noFCgiIhIUChQREQkKBQoIiISFP8fgNWf69Yim8cAAAAASUVORK5CYII=\n", "text/plain": [ "
" ] }, "metadata": { "tags": [] }, "output_type": "display_data" } ], "source": [ "learn.lr_find()\n", "learn.recorder.plot()" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 366 }, "colab_type": "code", "id": "MWrIKJCPTDuF", "outputId": "3545b480-c65b-4e10-e971-a54855166940" }, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
epochtrain_lossvalid_lossaccuracytime
00.5446820.5662840.74000000:00
10.4904250.4735220.84500000:00
20.4676820.4539770.84500000:00
30.4480430.4142920.83000000:00
40.4341180.4069610.86500000:00
50.4242440.4268440.85500000:00
60.4137500.4115790.84500000:00
" ], "text/plain": [ "" ] }, "metadata": { "tags": [] }, "output_type": "display_data" } ], "source": [ "#fit the learner\n", "learn.fit(7, 1e-2) #Number of epocs and the learning rate. learn.save('final_train')" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 221 }, "colab_type": "code", "id": "HrKFzLnGdTfN", "outputId": "a6f5fdf6-69ba-4ba5-db28-188351104eff" }, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PclassSexEmbarkedTitleFamilySAgeFareSibSpParchNameLengthtargetprediction
3maleSMr.Couple-0.6046-0.49820.5043-0.4658-0.416300
1femaleCMrs.Couple0.64330.71440.5043-0.46582.601711
3femaleSMiss.Single-0.2926-0.4855-0.4610-0.4658-0.524110
1femaleSMrs.Couple0.40930.37000.5043-0.46581.847211
3maleSMr.Single0.4093-0.4831-0.4610-0.4658-0.308500
" ], "text/plain": [ "" ] }, "metadata": { "tags": [] }, "output_type": "display_data" } ], "source": [ "#Show the results\n", "learn.show_results()" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 54 }, "colab_type": "code", "id": "uC533fQw8i3L", "outputId": "eb76b970-0633-447c-f423-a2115dfeda48" }, "outputs": [ { "data": { "text/plain": [ "(418,)" ] }, "execution_count": 11, "metadata": { "tags": [] }, "output_type": "execute_result" } ], "source": [ "#This will get predictions\n", "predictions, *_ = learn.get_preds(DatasetType.Test)\n", "labels = to_np(np.argmax(predictions, 1))\n", "labels.shape\n", "\n", "\n" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "7RpqfVNCa0V8" }, "outputs": [], "source": [ "#Writing to File\n", "submission=pd.DataFrame(test.loc[:,['PassengerId']])\n", "submission['Survived']=labels\n", "#Any files you save will be available in the output tab below\n", "\n", "submission.to_csv('submission.csv', index=False)" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "jGcsZ5ktaa-J" }, "outputs": [], "source": [ "from google.colab import files\n", "files.download('submission.csv')" ] } ], "metadata": { "accelerator": "GPU", "colab": { "collapsed_sections": [], "name": "01-titanic-fastai.ipynb", "provenance": [], "version": "0.3.2" }, "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": 1 }