{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "## Exercise 1: Getting familiar with pandas" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "collapsed": false }, "outputs": [], "source": [ "## Pandas \n", "# there are several ways to change a column in a dataframe\n", "# A short intro to pandas https://pandas.pydata.org/pandas-docs/stable/10min.html\n", "\n", "import pandas as pd\n", "import numpy as np\n", "import random" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# First step: make the data frame\n", "dates = pd.date_range('20130101', '20140101') #366\n", "data = pd.DataFrame(np.random.randn(366,4), index=dates, columns=list('ABCD'))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 1.1 Inspect the dataframe with the following commands: head(), tail(), describe." ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "collapsed": false }, "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", "
ABCD
2013-12-28-0.504931-0.1727683.0739181.418597
2013-12-292.187906-0.3891261.688600-1.812274
2013-12-30-0.600573-0.609973-0.6559191.431228
2013-12-31-1.408459-1.4219691.099690-1.036196
2014-01-010.894577-0.9968710.8169140.405218
\n", "
" ], "text/plain": [ " A B C D\n", "2013-12-28 -0.504931 -0.172768 3.073918 1.418597\n", "2013-12-29 2.187906 -0.389126 1.688600 -1.812274\n", "2013-12-30 -0.600573 -0.609973 -0.655919 1.431228\n", "2013-12-31 -1.408459 -1.421969 1.099690 -1.036196\n", "2014-01-01 0.894577 -0.996871 0.816914 0.405218" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Solution\n", "\n", "data.describe()\n", "data.head()\n", "data.tail()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 1.2 The index is a time series, and pandas has a build-in command for re-sampling dataframes (documentation: https://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.resample.html). Use resample to get the median every 2 days and save this as a new dataframe." ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "collapsed": true }, "outputs": [], "source": [ "#Solution: \n", " \n", "new_data = data.resample('2D').median()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 1.3 Inspect the new dataframe to see the difference in size compared to the inital dataframe." ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "(A 366\n", " B 366\n", " C 366\n", " D 366\n", " dtype: int64, A 183\n", " B 183\n", " C 183\n", " D 183\n", " dtype: int64)" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data.count(), new_data.count()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 1.4 Write your new dataframe to a csv file." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [ "new_data.to_csv('test_pythoncourse', sep='\\t')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 1.5 Merge the two dataframes. There are several ways to do this, see also https://pandas.pydata.org/pandas-docs/stable/merging.html." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "collapsed": false }, "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", "
ABCD
0-0.990819-0.500740-0.3784290.587768
1-0.8803900.642009-1.2549210.371295
2-0.921438-0.152966-0.5796761.027905
3-0.6314040.5837031.164311-0.049322
40.5077870.616022-0.300950-0.757167
\n", "
" ], "text/plain": [ " A B C D\n", "0 -0.990819 -0.500740 -0.378429 0.587768\n", "1 -0.880390 0.642009 -1.254921 0.371295\n", "2 -0.921438 -0.152966 -0.579676 1.027905\n", "3 -0.631404 0.583703 1.164311 -0.049322\n", "4 0.507787 0.616022 -0.300950 -0.757167" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Solution\n", "#generate another dataframe and merge them together\n", "new_data.merge(data, how='left').head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 1.6 There are several ways to perform actions on the dataframe columns. The dataframe has several columns containing negative values. For this exercise, find these negative values on a column, and create a new column with their absolute value, using a list comprehension, and after this, using a lambda function. You can use the magic timeit to see if there is a difference between these operations." ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The slowest run took 12.78 times longer than the fastest. This could mean that an intermediate result is being cached.\n", "1000 loops, best of 3: 226 µs per loop\n" ] } ], "source": [ "# Solution\n", "# method 1: list comprehension\n", "%timeit data['E'] = [ abs(x) for x in data['B'] ]" ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The slowest run took 4.87 times longer than the fastest. This could mean that an intermediate result is being cached.\n", "1000 loops, best of 3: 279 µs per loop\n" ] } ], "source": [ "# method 2: lambda function\n", "\n", "%timeit data['F'] = data['B'].apply(lambda x: abs(x) ) \n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Exercise 2: Supervised learning: Classification of MNIST data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 2.1 Download the digit ('MNIST original') dataset from mldata.org, which is a public repository for machine learning data. Divide the data into training and testing. Please use 1/7 for training and the rest for testing. \n", "\n", "Hint: The sklearn.datasets package is able to directly download data sets from the repository using the function sklearn.datasets.fetch_mldata. Generate the training and testing set by importing train_test_split from sklearn.model_selection\n", "\n", " " ] }, { "cell_type": "code", "execution_count": 22, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# Solution\n", "import sklearn \n", "from sklearn.datasets import fetch_mldata\n", "mnist = fetch_mldata('MNIST original')\n", "\n", "from sklearn.model_selection import train_test_split\n", "train_img, test_img, train_label, test_label = train_test_split(mnist.data, mnist.target, test_size=1/7.0, random_state=0)\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "collapsed": true }, "source": [ "### 2.2 The optimal performance of many machine learning algorithms is effected by scale. Typically, you need to scale the features in your data before applying any algorithm. Normalize the data and plot some random images from the dataset. \n", "\n", "Hint: Use StandardScaler from sklearn.preprocessing to help you standardize the dataset’s features onto unit scale (mean = 0 and variance = 1)\n", "\n" ] }, { "cell_type": "code", "execution_count": 23, "metadata": { "collapsed": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "//anaconda/lib/python3.5/site-packages/sklearn/utils/validation.py:444: DataConversionWarning: Data with input dtype uint8 was converted to float64 by StandardScaler.\n", " warnings.warn(msg, DataConversionWarning)\n" ] } ], "source": [ "# Solution\n", "from sklearn.preprocessing import StandardScaler\n", "scaler = StandardScaler()\n", "\n", "# Fit on training set only\n", "scaler.fit(train_img)\n", "\n", "# Apply transform to both the training set and the test set\n", "train_img = scaler.transform(train_img)\n", "test_img = scaler.transform(test_img)\n", "\n" ] }, { "cell_type": "code", "execution_count": 24, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABjkAAAFeCAYAAADXK32QAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Visualization\n", "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", "plt.figure(figsize=(20,4))\n", "for index, (image, label) in enumerate(zip(train_img[5:10], train_label[5:10])):\n", " plt.subplot(1, 5, index + 1)\n", " plt.imshow(np.reshape(image, (28,28)), cmap=plt.cm.gray)\n", " plt.title('Training example: %i\\n' % label, fontsize = 14)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 2.3 Logistic regression is one of the simplest linear classification algorithms. Fit a logistic regression model to the training images. Compute the accuracy of the classifier on the test images, and the time needed to train the model.¶\n", "\n", "Hint: Use LogisticRegression from sklearn.linear_model. To increase speed, change the default solver to 'lbfgs'\n" ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The total time is 42.05135893821716 seconds \n", "The classification accuracy is 0.9157 \n" ] } ], "source": [ "# Solution\n", "from time import time\n", "from sklearn.linear_model import LogisticRegression\n", "\n", "tic = time()\n", "logisticRegr = LogisticRegression(solver = 'lbfgs')\n", "logisticRegr.fit(train_img, train_label)\n", "score = logisticRegr.score(test_img, test_label)\n", "toc = time()\n", "print('The total time is %s seconds ' % (toc-tic))\n", "print('The classification accuracy is %s ' % score)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 2.4 Apply Principle Component Analysis (PCA) to the training signals by keeping only (a) 25%, (b) 75%, and (c) 95% of the energy. For each of the three cases, output the number of the required principle components.Then, plot the Cumulative Explained Variance over PCA. Finally, choose a random image from the dataset, and show its approximation with the PCA components. \n", "\n", "Hint: For computing the Cumulative Explained Variance over PCA use:\n", "```\n", "pca.explained_variance_ratio_.cumsum()\n", "\n", "```" ] }, { "cell_type": "code", "execution_count": 26, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "PCA used 330 components\n" ] } ], "source": [ "# Solution \n", "from sklearn.decomposition import PCA\n", "pca = PCA(.95)\n", "pca.fit(train_img)\n", "components = pca.transform(train_img)\n", "approximation = pca.inverse_transform(components)\n", "print('PCA used %s components' %pca.n_components_)" ] }, { "cell_type": "code", "execution_count": 29, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqgAAAG/CAYAAACUgyXiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plot the Cumulative Explained Variance over PCA\n", "\n", "plt.figure(figsize=(8, 5));\n", "plt.title('Cumulative Explained Variance over PCA');\n", "plt.plot(pca.explained_variance_ratio_.cumsum());" ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAApcAAAFXCAYAAAD3bGW4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSZIkFWNyKUmSpGL+PzN/IPre2DuRAAAAAElFTkSuQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Choose a random image from the dataset, and show its approximation with the PCA components\n", "\n", "plt.figure(figsize=(8,4));\n", "\n", "# Original Image\n", "plt.subplot(1, 2, 1);\n", "plt.imshow(train_img[0].reshape(28,28),\n", " cmap = plt.cm.gray, interpolation='nearest',\n", " clim=(0, 1));\n", "\n", "plt.title('Original Image', fontsize = 14);\n", "\n", "plt.subplot(1, 2, 2);\n", "plt.imshow(approximation[0].reshape(28, 28),\n", " cmap = plt.cm.gray, interpolation='nearest',\n", " clim=(0, 1));\n", "plt.title('95% of Explained Variance', fontsize = 14);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 2.5 Fit a logistic regression model to the approximation of the training images with 95% of explained variance. Compute the accuracy of the classifier and the time needed to train the model. Compare it to the one obtained in 2.3. What do you observe? \n" ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The total time is 24.675620079040527 seconds\n", "The classification accuracy is 0.9201 \n" ] } ], "source": [ "# Solution\n", "\n", "tic = time()\n", "logisticRegr = LogisticRegression(solver = 'lbfgs')\n", "logisticRegr.fit(components, train_label)\n", "components_test = pca.transform(test_img)\n", "score = logisticRegr.score(components_test, test_label)\n", "toc = time()\n", "\n", "print('The total time is %s seconds' % (toc-tic))\n", "print('The classification accuracy is %s ' % score)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Exercise 3: Unsupervised learning with Kmeans" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 3.1: Generate a set of 6 isotropic Gaussian blobs, with 1000 samples each. Each sample should have 60 features. \n", "\n", "Hint: Use the sklearn.datasets.make_blobs to generate the data" ] }, { "cell_type": "code", "execution_count": 34, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# Solution\n", "\n", "from sklearn.cluster import KMeans\n", "\n", "from sklearn.datasets.samples_generator import make_blobs\n", "X, y_true = make_blobs(n_samples=6000, n_features=60, centers=6,\n", " random_state=0)\n", "\n", "\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 3.2: Apply PCA to the generated data. Store the first two principle components and their cluster index to a new dataframe. Visualize the 6 blobs based only on these two components. " ] }, { "cell_type": "code", "execution_count": 42, "metadata": { "collapsed": false }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsQAAALBCAYAAABfmLcwAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "pca = PCA(n_components=2)\n", "components = pca.fit_transform(X)\n", "df = pd.DataFrame(data = components\n", " , columns = ['principal component 1', 'principal component 2'])\n", "df['cluster'] = y_true\n", "\n", "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", "\n", "fig = plt.figure(figsize = (8,8))\n", "ax = fig.add_subplot(1,1,1) \n", "ax.set_xlabel('Principal Component 1', fontsize = 12)\n", "ax.set_ylabel('Principal Component 2', fontsize = 12)\n", "ax.set_title('2 component PCA', fontsize = 12)\n", "targets = [0, 1, 2, 3, 4, 5]\n", "colors = ['r', 'g', 'b', 'w', 'm']\n", "for target, color in zip(targets,colors):\n", " indicesToKeep = df.cluster == target\n", " ax.scatter(df.loc[indicesToKeep, 'principal component 1']\n", " , df.loc[indicesToKeep, 'principal component 2']\n", " , c = color\n", " , s = 50)\n", "ax.legend(targets)\n", "ax.grid()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 3.3 Set the number of clusters to 6 and apply Kmeans clustering to the data. Compute the accuracy score between the true labels and the ones estimated by the Kmeans algorithm. " ] }, { "cell_type": "code", "execution_count": 38, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "1.0" ] }, "execution_count": 38, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Solution\n", "\n", "from sklearn.cluster import KMeans\n", "kmeans = KMeans(n_clusters=6)\n", "kmeans.fit(X)\n", "y_kmeans = kmeans.predict(X)\n", "\n", "\n", "from scipy.stats import mode\n", "\n", "labels = np.zeros_like(y_true)\n", "for i in range(6):\n", " mask = (y_kmeans == i)\n", " labels[mask] = mode(y_true[mask])[0]\n", " \n", "from sklearn.metrics import accuracy_score\n", "accuracy_score(y_true, labels)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 3.4 Do the same by clustering the data using only the first 2 principle components. What do you observe? " ] }, { "cell_type": "code", "execution_count": 40, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "1.0" ] }, "execution_count": 40, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Solution\n", "pca = PCA(n_components = 2)\n", "pca.fit(X)\n", "components = pca.transform(X)\n", "\n", "kmeans = KMeans(n_clusters=6)\n", "kmeans.fit(components)\n", "y_kmeans = kmeans.predict(components)\n", "\n", "labels = np.zeros_like(y_true)\n", "for i in range(6):\n", " mask = (y_kmeans == i)\n", " labels[mask] = mode(y_true[mask])[0]\n", " \n", "accuracy_score(y_true, labels)" ] } ], "metadata": { "anaconda-cloud": {}, "kernelspec": { "display_name": "Python [default]", "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": "3.5.2" } }, "nbformat": 4, "nbformat_minor": 2 }