{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Variational Autoencoders with Keras and MNIST\n", "Author: Charles Kenneth Fisher\n", "\n", "Adapted from: https://github.com/drckf/mlreview_notebooks/blob/master/jupyter_notebooks/notebooks/NB19_CXVII-Keras_VAE_MNIST.ipynb" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Learning Goals\n", "The goals of this notebook is to learn how to code a variational autoencoder in Keras. We will discuss hyperparameters, training, and loss-functions. In addition, we will familiarize ourselves with the Keras sequential GUI as well as how to visualize results and make predictions using a VAE with a small number of latent dimensions.\n", "\n", "## Overview\n", "\n", "This notebook teaches the reader how to build a Variational Autoencoder (VAE) with Keras. The code is a minimally modified, stripped-down version of the code from Lous Tiao in his wonderful [blog post](http://tiao.io/posts/implementing-variational-autoencoders-in-keras-beyond-the-quickstart-tutorial/) which the reader is strongly encouraged to also read.\n", "\n", "Our VAE will have Gaussian Latent variables and a Gaussian Posterior distribution $q_\\phi({\\mathbf z}|{\\mathbf x})$ with a diagonal covariance matrix. \n", "\n", "Recall, that a VAE consists of four essential elements:\n", "\n", "* A latent variable ${\\mathbf z}$ drawn from a distribution $p({\\mathbf z})$ which in our case will be a Gaussian with mean zero and standard\n", "deviation $\\epsilon$.\n", "* A decoder $p(\\mathbf{x}|\\mathbf{z})$ that maps latent variables ${\\mathbf z}$ to visible variables ${\\mathbf x}$. In our case, this is just a Multi-Layer Perceptron (MLP) - a neural network with one hidden layer.\n", "* An encoder $q_\\phi(\\mathbf{z}|\\mathbf{x})$ that maps examples to the latent space. In our case, this map is just a Gaussian with means and variances that depend on the input: $q_\\phi({\\bf z}|{\\bf x})= \\mathcal{N}({\\bf z}, \\boldsymbol{\\mu}({\\bf x}), \\mathrm{diag}(\\boldsymbol{\\sigma}^2({\\bf x})))$\n", "* A cost function consisting of two terms: the reconstruction error and an additional regularization term that minimizes the KL-divergence between the variational and true encoders. Mathematically, the reconstruction error is just the cross-entropy between the samples and their reconstructions. The KL-divergence term can be calculated analytically for this term and can be written as\n", "\n", "$$-D_{KL}(q_\\phi({\\bf z}|{\\bf x})|p({\\bf z}))={1 \\over 2} \\sum_{j=1}^J \\left (1+\\log{\\sigma_j^2({\\bf x})}-\\mu_j^2({\\bf x}) -\\sigma_j^2({\\bf x})\\right).\n", "$$\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Importing Data and specifying hyperparameters\n", "\n", "In the next section of code, we import the data and specify hyperparameters. The MNIST data are gray scale ranging in values from 0 to 255 for each pixel. We normalize this range to lie between 0 and 1. \n", "\n", "The hyperparameters we need to specify the architecture and train the VAE are:\n", "\n", "* The dimension of the hidden layers for encoders and decoders (`intermediate_dim`)\n", "* The dimension of the latent space (`latent_dim`)\n", "* The standard deviation of latent variables (`epsilon_std`)\n", "* Optimization hyper-parameters: `batch_size`, `epochs`\n" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(60000, 28, 28, 1)\n" ] } ], "source": [ "import numpy as np\n", "import matplotlib.pyplot as plt\n", "from scipy.stats import norm\n", "\n", "from keras import backend as K\n", "\n", "from keras.layers import (Input, InputLayer, Dense, Lambda, Layer, \n", " Add, Multiply)\n", "from keras.models import Model, Sequential\n", "from keras.datasets import mnist\n", "import pandas as pd\n", "\n", "\n", "#Load Data and map gray scale 256 to number between zero and 1\n", "(x_train, y_train), (x_test, y_test) = mnist.load_data()\n", "x_train = np.expand_dims(x_train, axis=-1) / 255.\n", "x_test = np.expand_dims(x_test, axis=-1) / 255.\n", "\n", "print(x_train.shape)\n", "\n", "# Find dimensions of input images\n", "img_rows, img_cols, img_chns = x_train.shape[1:]\n", "\n", "# Specify hyperparameters\n", "original_dim = img_rows * img_cols\n", "intermediate_dim = 256\n", "latent_dim = 2\n", "batch_size = 100\n", "epochs = 10\n", "epsilon_std = 1.0" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Specifying the loss function\n", "\n", "Here we specify the loss function. The first block of code is just the reconstruction error which is given by the cross-entropy. The second block of code calculates the KL-divergence analytically and adds it to the loss function with the line `self.add_loss`. It represents the KL-divergence as just another layer in the neural network with the inputs equal to the outputs: the means and variances for the variational encoder (i.e. $\\boldsymbol{\\mu}({\\bf x})$ and $\\boldsymbol{\\sigma}^2({\\bf x})$)." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "def nll(y_true, y_pred):\n", " \"\"\" Negative log likelihood (Bernoulli). \"\"\"\n", "\n", " # keras.losses.binary_crossentropy gives the mean\n", " # over the last axis. we require the sum\n", " return K.sum(K.binary_crossentropy(y_true, y_pred), axis=-1)\n", "\n", "class KLDivergenceLayer(Layer):\n", "\n", " \"\"\" Identity transform layer that adds KL divergence\n", " to the final model loss.\n", " \"\"\"\n", "\n", " def __init__(self, *args, **kwargs):\n", " self.is_placeholder = True\n", " super(KLDivergenceLayer, self).__init__(*args, **kwargs)\n", "\n", " def call(self, inputs):\n", "\n", " mu, log_var = inputs\n", "\n", " kl_batch = - .5 * K.sum(1 + log_var -\n", " K.square(mu) -\n", " K.exp(log_var), axis=-1)\n", "\n", " self.add_loss(K.mean(kl_batch), inputs=inputs)\n", "\n", " return inputs" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Encoder and Decoder\n", "\n", "The following specifies both the encoder and decoder. The encoder is a MLP with three layers that maps ${\\bf x}$ to $\\boldsymbol{\\mu}({\\bf x})$ and $\\boldsymbol{\\sigma}^2({\\bf x})$, followed by the generation of a latent variable using the reparametrization trick (see main text). The decoder is specified as a single sequential Keras layer.\n" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "WARNING:tensorflow:From /home/cms.woodson/miniconda3/envs/machine-learning-hats-2020/lib/python3.6/site-packages/tensorflow/python/ops/resource_variable_ops.py:435: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n", "Instructions for updating:\n", "Colocations handled automatically by placer.\n" ] } ], "source": [ "# Encoder\n", "\n", "x = Input(shape=(original_dim,))\n", "h = Dense(intermediate_dim, activation='relu')(x)\n", "\n", "z_mu = Dense(latent_dim)(h)\n", "z_log_var = Dense(latent_dim)(h)\n", "\n", "z_mu, z_log_var = KLDivergenceLayer()([z_mu, z_log_var])\n", "\n", "# Reparametrization trick\n", "z_sigma = Lambda(lambda t: K.exp(.5*t))(z_log_var)\n", "\n", "eps = Input(tensor=K.random_normal(shape=(K.shape(x)[0], \n", " latent_dim)))\n", "z_eps = Multiply()([z_sigma, eps])\n", "z = Add()([z_mu, z_eps])\n", "\n", "# This defines the Encoder which takes noise and input and outputs\n", "# the latent variable z\n", "encoder = Model(inputs=[x, eps], outputs=z)\n", "\n", "# Decoder is MLP specified as single Keras Sequential Layer\n", "decoder = Sequential([\n", " Dense(intermediate_dim, input_dim=latent_dim, activation='relu'),\n", " Dense(original_dim, activation='sigmoid')\n", "])\n", "\n", "x_pred = decoder(z)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Training the model\n", "\n", "We now train the model. Even though the loss function is the negative log likelihood (cross-entropy), recall that the KL-layer adds the analytic form of the loss function as well. We also have to reshape the data to make it a vector, and specify an optimizer." ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Train on 60000 samples, validate on 10000 samples\n", "Epoch 1/10\n", "60000/60000 [==============================] - 8s 137us/step - loss: 164.4510 - val_loss: 163.7581\n", "Epoch 2/10\n", "60000/60000 [==============================] - 8s 132us/step - loss: 162.9322 - val_loss: 162.6370\n", "Epoch 3/10\n", "60000/60000 [==============================] - 8s 138us/step - loss: 161.6371 - val_loss: 161.2954\n", "Epoch 4/10\n", "60000/60000 [==============================] - 8s 131us/step - loss: 160.4372 - val_loss: 160.2086\n", "Epoch 5/10\n", "60000/60000 [==============================] - 8s 136us/step - loss: 159.3402 - val_loss: 159.1020\n", "Epoch 6/10\n", "60000/60000 [==============================] - 8s 142us/step - loss: 158.3837 - val_loss: 158.5198\n", "Epoch 7/10\n", "60000/60000 [==============================] - 8s 141us/step - loss: 157.5569 - val_loss: 157.7762\n", "Epoch 8/10\n", "60000/60000 [==============================] - 8s 139us/step - loss: 156.8717 - val_loss: 157.2459\n", "Epoch 9/10\n", "60000/60000 [==============================] - 8s 136us/step - loss: 156.2878 - val_loss: 157.0014\n", "Epoch 10/10\n", "60000/60000 [==============================] - 8s 139us/step - loss: 155.7507 - val_loss: 155.9476\n" ] } ], "source": [ "vae = Model(inputs=[x, eps], outputs=x_pred, name='vae')\n", "vae.compile(optimizer='rmsprop', loss=nll)\n", "\n", "(x_train, y_train), (x_test, y_test) = mnist.load_data()\n", "x_train = x_train.reshape(-1, original_dim) / 255.\n", "x_test = x_test.reshape(-1, original_dim) / 255.\n", "\n", "hist = vae.fit(\n", " x_train,\n", " x_train,\n", " shuffle=True,\n", " epochs=epochs,\n", " batch_size=batch_size,\n", " validation_data=(x_test, x_test)\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Visualizing the loss function\n", "\n", "We can automatically visualize the loss function as a function of the epoch using the standard Keras interface for fitting. " ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "%matplotlib inline\n", "\n", "#for pretty plots\n", "golden_size = lambda width: (width, 2. * width / (1 + np.sqrt(5)))\n", "\n", "fig, ax = plt.subplots(figsize=golden_size(6))\n", "\n", "hist_df = pd.DataFrame(hist.history)\n", "hist_df.plot(ax=ax)\n", "\n", "ax.set_ylabel('NELBO')\n", "ax.set_xlabel('# epochs')\n", "\n", "ax.set_ylim(.99*hist_df[1:].values.min(), \n", " 1.1*hist_df[1:].values.max())\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Visualizing embedding in latent space\n", "\n", "Since our latent space is two dimensional, we can think of our encoder as defining a dimensional reduction of the original 784 dimensional space to just two dimensions! We can visualize the structure of this mapping by plotting the MNIST dataset in the latent space, with each point colored by which number it is $[0,1,\\ldots,9]$." ] }, { "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_test_encoded = encoder.predict(x_test, batch_size=batch_size)\n", "plt.figure(figsize=golden_size(6))\n", "plt.scatter(x_test_encoded[:, 0], x_test_encoded[:, 1], c=y_test, cmap='nipy_spectral')\n", "plt.colorbar()\n", "plt.savefig('VAE_MNIST_latent.pdf')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Generating new examples\n", "\n", "One of the nice things about VAEs is that they are generative models. Thus, we can generate new examples.\n", "\n", "* Sampling uniformally in the latent space \n", "* Sampling accounting for the fact that the latent space is Gaussian so that we expect most of the data points to be centered around (0,0) and fall off exponentially in all directions. This is done by transforming the uniform grid using the inverse Cumulative Distribution Function (CDF) for the Gaussian.\n", "\n" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# display a 2D manifold of the images\n", "n = 5 # figure with 15x15 images\n", "quantile_min = 0.01\n", "quantile_max = 0.99\n", "\n", "# Linear Sampling\n", "# we will sample n points within [-15, 15] standard deviations\n", "z1_u = np.linspace(5, -5, n)\n", "z2_u = np.linspace(5, -5, n)\n", "z_grid = np.dstack(np.meshgrid(z1_u, z2_u))\n", "\n", "x_pred_grid = decoder.predict(z_grid.reshape(n*n, latent_dim)) \\\n", " .reshape(n, n, img_rows, img_cols)\n", "\n", "# Plot figure\n", "fig, ax = plt.subplots(figsize=golden_size(10))\n", "\n", "ax.imshow(np.block(list(map(list, x_pred_grid))), cmap='gray')\n", "\n", "ax.set_xticks(np.arange(0, n*img_rows, img_rows) + .5 * img_rows)\n", "ax.set_xticklabels(map('{:.2f}'.format, z1_u), rotation=90)\n", "\n", "ax.set_yticks(np.arange(0, n*img_cols, img_cols) + .5 * img_cols)\n", "ax.set_yticklabels(map('{:.2f}'.format, z2_u))\n", "\n", "ax.set_xlabel('$z_1$')\n", "ax.set_ylabel('$z_2$')\n", "ax.set_title('Uniform')\n", "ax.grid(False)\n", "\n", "\n", "plt.savefig('VAE_MNIST_fantasy_uniform.pdf')\n", "plt.show()\n", "\n", "# Inverse CDF sampling\n", "z1 = norm.ppf(np.linspace(quantile_min, quantile_max, n))\n", "z2 = norm.ppf(np.linspace(quantile_max, quantile_min, n))\n", "z_grid2 = np.dstack(np.meshgrid(z1, z2))\n", "\n", "x_pred_grid2 = decoder.predict(z_grid2.reshape(n*n, latent_dim)) \\\n", " .reshape(n, n, img_rows, img_cols)\n", "\n", "# Plot figure Inverse CDF sampling\n", "fig, ax = plt.subplots(figsize=golden_size(10))\n", "\n", "ax.imshow(np.block(list(map(list, x_pred_grid2))), cmap='gray')\n", "\n", "ax.set_xticks(np.arange(0, n*img_rows, img_rows) + .5 * img_rows)\n", "ax.set_xticklabels(map('{:.2f}'.format, z1), rotation=90)\n", "\n", "ax.set_yticks(np.arange(0, n*img_cols, img_cols) + .5 * img_cols)\n", "ax.set_yticklabels(map('{:.2f}'.format, z2))\n", "\n", "ax.set_xlabel('$z_1$')\n", "ax.set_ylabel('$z_2$')\n", "ax.set_title('Inverse CDF')\n", "ax.grid(False)\n", "plt.savefig('VAE_MNIST_fantasy_invCDF.pdf')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Exercises\n", "\n", "* Play with the standard deviation of the latent variables $\\epsilon$. How does this effect your results?\n", "* Generate samples as you increase the number of latent dimensions. Do your generated samples look better? Visualize the latent variables using a dimensional reduction technique such as PCA or t-SNE. How does it compare to the case with two latent dimensions showed above?\n", "* Repeat this analysis with the W tagging dataset." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "machine-learning-hats-2020", "language": "python", "name": "machine-learning-hats-2020" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.10" }, "nikola": { "category": "", "date": "2017-07-21 18:38:07 UTC+10:00", "description": "", "link": "", "slug": "variational-inference-with-implicit-approximate-inference-models-wip-pt-8", "tags": "", "title": "Variational Inference with Implicit Approximate Inference Models (WIP Pt. 8)", "type": "text" } }, "nbformat": 4, "nbformat_minor": 2 }