{ "cells": [ { "cell_type": "markdown", "metadata": { "run_control": { "marked": true } }, "source": [ "# Naive Bayes" ] }, { "cell_type": "markdown", "metadata": { "run_control": { "marked": true } }, "source": [ "## Introduction\n", "\n", "We will start by working on the Iris dataset. Recall that Iris dataset contains iris species and sepal and petal measurements. We will quickly explore the dataset and jump into Naive Bayes." ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "run_control": { "marked": true } }, "outputs": [], "source": [ "from __future__ import print_function\n", "import os\n", "#Data Path has to be set as per the file location in your system\n", "#data_path = ['..', 'data']\n", "data_path = ['data']" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Question 1\n", "\n", "* Load the Iris dataset.\n", "* Take a quick look at the data types.\n", "* Look at the skew values and decide if any transformations need to be applied. You can use skew value 0.75 as a threshold.\n", "* Use `sns.pairplot` to plot the pairwise correlations and histograms. Use `hue=\"species\"` as a keyword argument in order to see the distribution of species." ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "run_control": { "marked": true } }, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "#The filepath is dependent on the data_path set in the previous cell \n", "filepath = os.sep.join(data_path + ['Iris_Data.csv'])\n", "data = pd.read_csv(filepath, sep=',', header=0)" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "sepal_length float64\n", "sepal_width float64\n", "petal_length float64\n", "petal_width float64\n", "species object\n", "dtype: object" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.dtypes" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Notice that aside from the predictor variable, everything is float." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
skewtoo_skewed
sepal_length0.314911False
sepal_width0.334053False
petal_length-0.274464False
petal_width-0.104997False
\n", "
" ], "text/plain": [ " skew too_skewed\n", "sepal_length 0.314911 False\n", "sepal_width 0.334053 False\n", "petal_length -0.274464 False\n", "petal_width -0.104997 False" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "skew = pd.DataFrame(data.skew())\n", "skew.columns = ['skew']\n", "skew['too_skewed'] = skew['skew'] > .75\n", "skew" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Fields are not too badly skewed." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline\n", "sns.pairplot(data, hue='species')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Question 2\n", "\n", "Let's now fit a Naive Bayes classifier to this data in order to predict \"species\".\n", "\n", "* Pick the appropriate type of Naive Bayes given the nature of your dataset (data types of columns). Recall, choices are\n", " * GaussianNB\n", " * MultinomialNB\n", " * BernoulliNB\n", "* Use `cross_val_score` to see how well your choice works." ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[0.94871795 0.94871795 0.91666667 1. ]\n" ] }, { "data": { "text/plain": [ "0.953525641025641" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Since the features are continuous, the right choice is GaussianNB\n", "\n", "from sklearn.naive_bayes import GaussianNB\n", "from sklearn.model_selection import cross_val_score\n", "X = data[data.columns[:-1]]\n", "y = data.species\n", "\n", "GNB = GaussianNB()\n", "cv_N = 4\n", "scores = cross_val_score(GNB, X, y, n_jobs=cv_N, cv=cv_N)\n", "print(scores)\n", "np.mean(scores)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Question 3:\n", "\n", "Now let's try all types of Naive Bayes and observe what happens\n", "\n", "* Compare the cross validation scores for Gaussian, Bernouilli and Multinomial Naive Bayes.\n", "* Why is BernoulliNB performing like it does?" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'gaussian': 0.953525641025641,\n", " 'bernoulli': 0.3333333333333333,\n", " 'multinomial': 0.9529914529914529}" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.naive_bayes import GaussianNB, BernoulliNB, MultinomialNB\n", "X = data[data.columns[:-1]]\n", "y = data.species\n", "nb = {'gaussian': GaussianNB(),\n", " 'bernoulli': BernoulliNB(),\n", " 'multinomial': MultinomialNB()}\n", "scores = {}\n", "for key, model in nb.items():\n", " s = cross_val_score(model, X, y, cv=cv_N, n_jobs=cv_N, scoring='accuracy')\n", " scores[key] = np.mean(s)\n", "scores" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Looks like BernoulliNB results are very bad, but MultinomialNB is doing a very good job.\n", "\n", "Why are the results of Bernoulli bad? Find out the reason." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Question 4:\n", "\n", "Let's see what happens when we take away the predictive features.\n", "\n", "* Check the pairplot histograms (diagonal) you produced above and identify the two most predictive features visually.\n", "* Remove the *petal_* features which are very predictive, and re-do the comparison above. That is, get the cross validation scores for all types of Naive Bayes." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "X = data[['sepal_length', 'sepal_width']]\n", "y = data.species\n", "\n", "nb = {'gaussian': GaussianNB(),\n", " 'bernoulli': BernoulliNB(),\n", " 'multinomial': MultinomialNB()}\n", "\n", "# Try other variants on the lines shown in the previous cell for GaussianNB and compare the results on scoring = 'accuracy'. \n", "# Run the piece of code as shown in array in question 3\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#Come up with your observations after taking away the very predictive features, which model works better" ] }, { "cell_type": "markdown", "metadata": { "run_control": { "marked": true } }, "source": [ "## Question 5\n", "\n", "This question explores how Naive Bayes algorithms can be affected when we push the underlying (naive) assumption too much. Recall that the naive assumption is that the features in the training set are *independent* from each other.\n", "\n", "* Create **0, 1, 3, 5, 10, 50, 100** copies of `sepal_length` and fit a `GaussianNB` for each one.\n", "* Keep track of the save the average `cross_val_score`.\n", "* Create a plot of the saved scores over the number of copies." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "X = data[data.columns[:-1]]\n", "y = data.species\n", "\n", "n_copies = [0, 1, 3, 5, 10, 50, 100]\n", "\n", "\n", "def create_copies_sepal_length(X, n):\n", " X_new = X.copy()\n", " for i in range(n):\n", " X_new['sepal_length_copy%s' % i] = X['sepal_length']\n", " return X_new\n", "\n", "\n", "def get_cross_val_score(n):\n", " X_new = create_copies_sepal_length(X, n)\n", " scores = cross_val_score(GaussianNB(), X_new, y, cv=cv_N, n_jobs=cv_N)\n", " return np.mean(scores)\n", "\n", "\n", "avg_scores = pd.Series(\n", " [get_cross_val_score(n) for n in n_copies],\n", " index=n_copies)\n", "\n", "ax = avg_scores.plot()\n", "ax.set(\n", " xlabel='number of extra copies of \"sepal_length\"',\n", " ylabel='average accuracy score',\n", " title='Decline in Naive Bayes performance');\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Question 6 - Naive Bayes on Human Activity Recongnition\n", "\n", "In this question, we'll explore discretizing the dataset and then fitting MultinomialNB. \n", "\n", "* Load the Human Activity Recognition dataset. \n", "* Look at the data types. It's all continuous except for the target.\n", "* Create `X` and `y` from `data`. `y` is the \"Activity\" column.\n", "* Create training and test splits.\n", "* Fit a GaussianNB to the training split.\n", "* Get predictions on the test set.\n", "* use `sns.heatmap` to plot the confusion matrix for predictions." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "path = 'data/Human_Activity_Recognition_Using_Smartphones_Data.csv'\n", "data = pd.read_csv(path, sep=',')" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "float64 561\n", "object 1\n", "dtype: int64" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.dtypes.value_counts()" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "scrolled": true }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
tBodyAcc-mean()-XtBodyAcc-mean()-YtBodyAcc-mean()-ZtBodyAcc-std()-XtBodyAcc-std()-YtBodyAcc-std()-ZtBodyAcc-mad()-XtBodyAcc-mad()-YtBodyAcc-mad()-ZtBodyAcc-max()-X...fBodyBodyGyroJerkMag-skewness()fBodyBodyGyroJerkMag-kurtosis()angle(tBodyAccMean,gravity)angle(tBodyAccJerkMean),gravityMean)angle(tBodyGyroMean,gravityMean)angle(tBodyGyroJerkMean,gravityMean)angle(X,gravityMean)angle(Y,gravityMean)angle(Z,gravityMean)Activity
00.288585-0.020294-0.132905-0.995279-0.983111-0.913526-0.995112-0.983185-0.923527-0.934724...-0.298676-0.710304-0.1127540.030400-0.464761-0.018446-0.8412470.179941-0.058627STANDING
10.278419-0.016411-0.123520-0.998245-0.975300-0.960322-0.998807-0.974914-0.957686-0.943068...-0.595051-0.8614990.053477-0.007435-0.7326260.703511-0.8447880.180289-0.054317STANDING
20.279653-0.019467-0.113462-0.995380-0.967187-0.978944-0.996520-0.963668-0.977469-0.938692...-0.390748-0.760104-0.1185590.1778990.1006990.808529-0.8489330.180637-0.049118STANDING
30.279174-0.026201-0.123283-0.996091-0.983403-0.990675-0.997099-0.982750-0.989302-0.938692...-0.117290-0.482845-0.036788-0.0128920.640011-0.485366-0.8486490.181935-0.047663STANDING
40.276629-0.016570-0.115362-0.998139-0.980817-0.990482-0.998321-0.979672-0.990441-0.942469...-0.351471-0.6992050.1233200.1225420.693578-0.615971-0.8478650.185151-0.043892STANDING
\n", "

5 rows × 562 columns

\n", "
" ], "text/plain": [ " tBodyAcc-mean()-X tBodyAcc-mean()-Y tBodyAcc-mean()-Z tBodyAcc-std()-X \\\n", "0 0.288585 -0.020294 -0.132905 -0.995279 \n", "1 0.278419 -0.016411 -0.123520 -0.998245 \n", "2 0.279653 -0.019467 -0.113462 -0.995380 \n", "3 0.279174 -0.026201 -0.123283 -0.996091 \n", "4 0.276629 -0.016570 -0.115362 -0.998139 \n", "\n", " tBodyAcc-std()-Y tBodyAcc-std()-Z tBodyAcc-mad()-X tBodyAcc-mad()-Y \\\n", "0 -0.983111 -0.913526 -0.995112 -0.983185 \n", "1 -0.975300 -0.960322 -0.998807 -0.974914 \n", "2 -0.967187 -0.978944 -0.996520 -0.963668 \n", "3 -0.983403 -0.990675 -0.997099 -0.982750 \n", "4 -0.980817 -0.990482 -0.998321 -0.979672 \n", "\n", " tBodyAcc-mad()-Z tBodyAcc-max()-X ... fBodyBodyGyroJerkMag-skewness() \\\n", "0 -0.923527 -0.934724 ... -0.298676 \n", "1 -0.957686 -0.943068 ... -0.595051 \n", "2 -0.977469 -0.938692 ... -0.390748 \n", "3 -0.989302 -0.938692 ... -0.117290 \n", "4 -0.990441 -0.942469 ... -0.351471 \n", "\n", " fBodyBodyGyroJerkMag-kurtosis() angle(tBodyAccMean,gravity) \\\n", "0 -0.710304 -0.112754 \n", "1 -0.861499 0.053477 \n", "2 -0.760104 -0.118559 \n", "3 -0.482845 -0.036788 \n", "4 -0.699205 0.123320 \n", "\n", " angle(tBodyAccJerkMean),gravityMean) angle(tBodyGyroMean,gravityMean) \\\n", "0 0.030400 -0.464761 \n", "1 -0.007435 -0.732626 \n", "2 0.177899 0.100699 \n", "3 -0.012892 0.640011 \n", "4 0.122542 0.693578 \n", "\n", " angle(tBodyGyroJerkMean,gravityMean) angle(X,gravityMean) \\\n", "0 -0.018446 -0.841247 \n", "1 0.703511 -0.844788 \n", "2 0.808529 -0.848933 \n", "3 -0.485366 -0.848649 \n", "4 -0.615971 -0.847865 \n", "\n", " angle(Y,gravityMean) angle(Z,gravityMean) Activity \n", "0 0.179941 -0.058627 STANDING \n", "1 0.180289 -0.054317 STANDING \n", "2 0.180637 -0.049118 STANDING \n", "3 0.181935 -0.047663 STANDING \n", "4 0.185151 -0.043892 STANDING \n", "\n", "[5 rows x 562 columns]" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.head()" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "from sklearn.model_selection import train_test_split\n", "\n", "X = data.drop('Activity', axis=1)\n", "y = data.Activity\n", "X_train, X_test, y_train, y_test = train_test_split(X,y)" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "scrolled": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(7724, 561)\n", "(2575, 561)\n" ] } ], "source": [ "print(X_train.shape)\n", "print(X_test.shape)" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.naive_bayes import GaussianNB\n", "from sklearn.metrics import confusion_matrix\n", "import seaborn as sns\n", "\n", "gnb = GaussianNB()\n", "gnb = gnb.fit(X_train, y_train)\n", "y_test_pred = gnb.predict(X_test)\n", "labels = sorted(y_test.unique())\n", "\n", "cm = pd.DataFrame(confusion_matrix(y_test, y_test_pred, labels), index=labels, columns=labels)\n", "sns.heatmap(cm, cmap='afmhot', xticklabels=True, yticklabels=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Question 7\n", "\n", "Now, let's discretize the dataset from Question 6. There are many ways to do this, but we'll use `pd.DataFrame.rank(pct=True)`.\n", "\n", "a. Create `X_discrete` from `X` using .rank(pct=True)\n", "\n", "b. Look at the values. They are still not discrete. Modify `X_discrete` so that it is indeed discrete. (Hint: try to get the first 2 digits using `.applymap`)\n", "\n", "c. Split `X_discrete` and `y` into training and test datasets\n", "\n", "d. Fit a MultinomialNB to the training split.\n", "\n", "e. Get predictions on the test set.\n", "\n", "f. Plot the confusion matrix for predictions." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "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": "3.7.3" }, "varInspector": { "cols": { "lenName": 16, "lenType": 16, "lenVar": 40 }, "kernels_config": { "python": { "delete_cmd_postfix": "", "delete_cmd_prefix": "del ", "library": "var_list.py", "varRefreshCmd": "print(var_dic_list())" }, "r": { "delete_cmd_postfix": ") ", "delete_cmd_prefix": "rm(", "library": "var_list.r", "varRefreshCmd": "cat(var_dic_list()) " } }, "types_to_exclude": [ "module", "function", "builtin_function_or_method", "instance", "_Feature" ], "window_display": false } }, "nbformat": 4, "nbformat_minor": 2 }