{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# 5.1 Explanation Continuity" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Introduction\n", "\n", "**This section corresponds to Section 7.1 of the original paper.**\n", "\n", "What characterizes a good explanation? A first desirable property of an explanation technique is that if it is given two similar images, then its explanation must also be similar. Explanation continuity (or lack of it) can be quantified by looking for the strongest variation of the explanation $R(x)$ in the input domain:\n", "\n", "\\begin{equation}\n", "\\max_{x \\neq x'} \\frac{\\lVert R(x) - R(x') \\lVert_1}{\\lVert x - x' \\lVert_2}\n", "\\end{equation}\n", "\n", "When $f(x)$ is a deep ReLU network, both sensitivity analysis and Simple Taylor Decomposition have sharp discontinuities in their explanation function while Deep Taylor Decomposition provides continuous explanations. Let us explore what this means with a simple function $f(\\mathbf{x}) = \\max (x_1, x_2)$ in $\\mathbb{R}_2^+$, here implemented by the two-layer ReLU network\n", "\n", "\\begin{equation}\n", "f(\\mathbf{x}) = \\max \\left( 0, \\frac{\\max (0, x_1 - x_2) + \\max (0, x_2 - x_1) + \\max (0, x_1 + x_2)}{2} \\right)\n", "\\end{equation}\n", "\n", "![title](./assets/5_1_EC/net.jpg)\n", "\n", "The network is comprised of three layers (input, hidden and output). The number at each line connecting the neurons indicates the associated weight value and the bent line denotes the use of ReLU activation function. $a_j$ and $b_k$ denote activations of the hidden and output layers respectively. Then, we can define the network as:\n", "\n", "\\begin{align}\n", "a_1 & = \\max (0, x_1 - x_2) \\\\\n", "a_2 & = \\max (0, x_2 - x_1) \\\\\n", "a_3 & = \\max (0, x_1 + x_2) = x_1 + x_2 \\\\\n", "b_1 & = \\max \\left( 0, \\frac{a_1 + a_2 + a_3}{2} \\right) = \\frac{a_1 + a_2 + a_3}{2} = \\max (x_1, x_2)\n", "\\end{align}\n", "We could remove the $\\max$ in *Eqs. (5)* and *(6)* because $x_1$ and $x_2$ are guaranteed to be positive.\n", "\n", "\n", "Now, we are ready to compare the various explanation techniques we convered throughout the tutorials. I will not calculate the relevance scores for Sensitivity Analysis and Simple Taylor Decomposition because they can be easily derived.\n", "\n", "### Sensitivity Analysis\n", "\n", "\\begin{equation}\n", "R(\\mathbf{x}) =\n", "\\begin{cases}\n", "(1,0) & \\text{if} \\ x_1 > x_2 \\\\\n", "(0,1) & \\text{if} \\ x_1 < x_2\n", "\\end{cases}\n", "\\end{equation}\n", "\n", "### Simple Taylor Decomposition\n", "\n", "\\begin{equation}\n", "R(\\mathbf{x}) =\n", "\\begin{cases}\n", "(x_1,0) & \\text{if} \\ x_1 > x_2 \\\\\n", "(0,x_2) & \\text{if} \\ x_1 < x_2\n", "\\end{cases}\n", "\\end{equation}\n", "\n", "### Deep Taylor Decomposition\n", "\n", "To derive the relevance at the input layer, we must backpropagate the function output throughout the layers using the Deep Taylor Decomposition rules. Since the input is constrained to be positive ($\\mathbb{R}_2^+$) and we are using ReLU activation in rest of the layers, we only need to apply the $z^+$-rule.\n", "\n", "#### Relevance propagation from output to hidden layer\n", "\n", "The $z^+$-rule is given below:\n", "\n", "\\begin{equation}\n", "R_{j \\leftarrow k}^{(2,3)} = \\frac{z_{jk}^+}{\\sum_k z_{jk}^+} R_{k}^{(3)}\n", "\\end{equation}\n", "where $z_{jk}^+ = a_j w_{jk}^+$ and where $w_{jk}^+$ denotes the positive part of $w_{jk}$.\n", "\n", "By the *Layer-wise Relevance Conservation Constraint*, $R_k^{(3)} = b_1$. $\\{z_{jk}^+\\}$ and $\\sum_j \\{z_{jk}^+\\}$ is given as follows:\n", "\n", "\\begin{equation}\n", "\\{z_{jk}^+\\} = \n", "\\begin{bmatrix}\n", "\\displaystyle \\frac{a_1}{2} \\\\ \\displaystyle \\frac{a_2}{2} \\\\ \\displaystyle \\frac{a_3}{2}\n", "\\end{bmatrix}\n", "\\end{equation}\n", "\n", "\\begin{equation}\n", "\\sum_j \\{z_{jk}^+\\} = \n", "\\begin{bmatrix}\n", "\\displaystyle \\frac{\\sum_j a_j}{2}\n", "\\end{bmatrix}\n", "\\end{equation}\n", "\n", "Then $R_{j \\leftarrow k}^{(2,3)} = R_{j}^{(2)}$ can be calculated with *Eq. (9)*.\n", "\n", "\\begin{equation}\n", "R_{j}^{(2)} = \n", "\\begin{bmatrix}\n", "\\displaystyle \\frac{a_1}{\\sum_j a_j} \\cdot b_1 \\\\ \\displaystyle \\frac{a_2}{\\sum_j a_j} \\cdot b_1 \\\\ \\displaystyle \\frac{a_3}{\\sum_j a_j} \\cdot b_1\n", "\\end{bmatrix}\n", "\\end{equation}\n", "\n", "#### Relevance propagation from hidden to input layer\n", "\n", "The $z^+$-rule is given below:\n", "\n", "\\begin{equation}\n", "R_{i \\leftarrow j}^{(1,2)} = \\frac{z_{ij}^+}{\\sum_j z_{ij}^+} R_{j}^{(2)}\n", "\\end{equation}\n", "where $z_{ij}^+ = x_i w_{ij}^+$ and where $w_{ij}^+$ denotes the positive part of $w_{ij}$.\n", "\n", "$\\{z_{ij}^+\\}$ and $\\sum_i \\{z_{ij}^+\\}$ is given as follows:\n", "\n", "\\begin{equation}\n", "\\{z_{ij}^+\\} = \n", "\\begin{bmatrix}\n", "x_1 & 0 & x_1 \\\\\n", "0 & x_2 & x_2\n", "\\end{bmatrix}\n", "\\end{equation}\n", "\n", "\\begin{equation}\n", "\\sum_i \\{z_{ij}^+\\} = \n", "\\begin{bmatrix}\n", "x_1 & x_2 & x_1 + x_2 \\\\\n", "\\end{bmatrix}\n", "\\end{equation}\n", "\n", "Then $R_{i}^{(1)} = \\sum_j R_{i \\leftarrow j}^{(1,2)}$ can be calculated with *Eq. (13)*.\n", "\n", "\\begin{equation}\n", "R_{i}^{(1)} = \n", "\\begin{bmatrix}\n", "\\displaystyle \\frac{a_1}{\\sum_j a_j} \\cdot b_1 + \\frac{x_1}{x_1+x_2} \\cdot \\frac{a_3}{\\sum_j a_j} \\cdot b_1 \\\\\n", "\\displaystyle \\frac{a_2}{\\sum_j a_j} \\cdot b_1 + \\frac{x_2}{x_1+x_2} \\cdot \\frac{a_3}{\\sum_j a_j} \\cdot b_1\n", "\\end{bmatrix} =\n", "\\begin{bmatrix}\n", "\\displaystyle \\frac{a_1}{2} + \\frac{x_1}{x_1+x_2} \\cdot \\frac{a_3}{2} \\\\\n", "\\displaystyle \\frac{a_2}{2} + \\frac{x_2}{x_1+x_2} \\cdot \\frac{a_3}{2}\n", "\\end{bmatrix} =\n", "\\begin{bmatrix}\n", "\\displaystyle \\frac{a_1 + x_1}{2} \\\\\n", "\\displaystyle \\frac{a_2 + x_2}{2}\n", "\\end{bmatrix}\n", "\\end{equation}\n", "\n", "This gives us the following relevance scores:\n", "\n", "\\begin{equation}\n", "R(\\mathbf{x}) =\n", "\\begin{cases}\n", "\\displaystyle \\left( \\frac{2 x_1 - x_2}{2}, \\frac{x_2}{2} \\right) & \\text{if} \\ x_1 > x_2 \\\\[1.5ex]\n", "\\displaystyle \\left( \\frac{x_1}{2}, \\frac{- x_1 + 2 x_2}{2} \\right) & \\text{if} \\ x_1 < x_2\n", "\\end{cases}\n", "\\end{equation}\n", "\n", "Visualization of all three techniques gives us the following figure:\n", "\n", "![title](./assets/5_1_EC/fig1.png)\n", "\n", "It can be seen that despite the continuity of the prediction function, the explanations offered by Sensitivity Analysis and Simple Taylor Decomposition are discontinuous on the line $x_1 = x_2$. Only Deep Taylor Decomposition produces a smooth transition. In the Tensorflow walkthrough, we are going to demonstrate the continuity of explanation techniques in another way. We are first going to train a DNN to recognize MNIST digits. Then, we are going to compare explanations offered by different explanation techniques as we move a handwritten digit from left to right. If the technique satisfies explanation continuity, the explanations produced also must change continuously without sudden jumps." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Tensorflow Walkthrough" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 1. Import Dependencies" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Extracting MNIST_data/train-images-idx3-ubyte.gz\n", "Extracting MNIST_data/train-labels-idx1-ubyte.gz\n", "Extracting MNIST_data/t10k-images-idx3-ubyte.gz\n", "Extracting MNIST_data/t10k-labels-idx1-ubyte.gz\n" ] } ], "source": [ "import os\n", "\n", "from tensorflow.examples.tutorials.mnist import input_data\n", "from tensorflow.python.ops import nn_ops, gen_nn_ops\n", "import matplotlib.pyplot as plt\n", "import matplotlib\n", "import tensorflow as tf\n", "import numpy as np\n", "\n", "from models.models_5_1 import MNIST_CNN, Taylor\n", "from utils import translate\n", "\n", "%matplotlib inline\n", "\n", "mnist = input_data.read_data_sets(\"MNIST_data/\", one_hot=True)\n", "images = mnist.train.images\n", "labels = mnist.train.labels\n", "\n", "logdir = './tf_logs/5_1_EQ/'\n", "ckptdir = logdir + 'model'\n", "\n", "if not os.path.exists(logdir):\n", " os.mkdir(logdir)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 2. Building Graph" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "collapsed": true }, "outputs": [], "source": [ "with tf.name_scope('Classifier'):\n", "\n", " # Initialize neural network\n", " DNN = MNIST_CNN('CNN')\n", "\n", " # Setup training process\n", " X = tf.placeholder(tf.float32, [None, 784], name='X')\n", " Y = tf.placeholder(tf.float32, [None, 10], name='Y')\n", "\n", " activations, logits = DNN(X)\n", " \n", " tf.add_to_collection('EC', X)\n", " \n", " for activation in activations:\n", " tf.add_to_collection('EC', activation)\n", "\n", " cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=logits, labels=Y))\n", "\n", " optimizer = tf.train.AdamOptimizer().minimize(cost, var_list=DNN.vars)\n", "\n", " correct_prediction = tf.equal(tf.argmax(logits, 1), tf.argmax(Y, 1))\n", " accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))\n", "\n", "cost_summary = tf.summary.scalar('Cost', cost)\n", "accuray_summary = tf.summary.scalar('Accuracy', accuracy)\n", "summary = tf.summary.merge_all()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 3. Training Network\n", "\n", "This is the step where the DNN is trained to classify the 10 digits of the MNIST images. Summaries are written into the logdir and you can visualize the statistics using tensorboard by typing this command: `tensorboard --lodir=./tf_logs`" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch: 0001 cost = 0.229207651 accuracy = 0.926581823\n", "Epoch: 0002 cost = 0.065176102 accuracy = 0.980072739\n", "Epoch: 0003 cost = 0.044646955 accuracy = 0.986000010\n", "Epoch: 0004 cost = 0.035834330 accuracy = 0.988909100\n", "Epoch: 0005 cost = 0.028319894 accuracy = 0.990672736\n", "Epoch: 0006 cost = 0.023727389 accuracy = 0.992454552\n", "Epoch: 0007 cost = 0.018540170 accuracy = 0.994145460\n", "Epoch: 0008 cost = 0.017687961 accuracy = 0.993981824\n", "Epoch: 0009 cost = 0.014346573 accuracy = 0.995436368\n", "Epoch: 0010 cost = 0.012755402 accuracy = 0.996072731\n", "Epoch: 0011 cost = 0.010847591 accuracy = 0.996745458\n", "Epoch: 0012 cost = 0.009276978 accuracy = 0.996854548\n", "Epoch: 0013 cost = 0.009246885 accuracy = 0.996890912\n", "Epoch: 0014 cost = 0.007504274 accuracy = 0.997436366\n", "Epoch: 0015 cost = 0.008067215 accuracy = 0.997290912\n", "Accuracy: 0.9935\n" ] } ], "source": [ "sess = tf.InteractiveSession()\n", "sess.run(tf.global_variables_initializer())\n", "\n", "saver = tf.train.Saver()\n", "file_writer = tf.summary.FileWriter(logdir, tf.get_default_graph())\n", "\n", "# Hyper parameters\n", "training_epochs = 15\n", "batch_size = 100\n", "\n", "for epoch in range(training_epochs):\n", " total_batch = int(mnist.train.num_examples / batch_size)\n", " avg_cost = 0\n", " avg_acc = 0\n", " \n", " for i in range(total_batch):\n", " batch_xs, batch_ys = mnist.train.next_batch(batch_size)\n", " _, c, a, summary_str = sess.run([optimizer, cost, accuracy, summary], feed_dict={X: batch_xs, Y: batch_ys})\n", " avg_cost += c / total_batch\n", " avg_acc += a / total_batch\n", " \n", " file_writer.add_summary(summary_str, epoch * total_batch + i)\n", "\n", " print('Epoch:', '%04d' % (epoch + 1), 'cost =', '{:.9f}'.format(avg_cost), 'accuracy =', '{:.9f}'.format(avg_acc))\n", " \n", " saver.save(sess, ckptdir)\n", "\n", "print('Accuracy:', sess.run(accuracy, feed_dict={X: mnist.test.images, Y: mnist.test.labels}))\n", "\n", "sess.close()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 4. Restoring Subgraph\n", "\n", "Here we first rebuild the DNN graph from metagraph, restore DNN parameters from the checkpoint and then gather the necessary nodes using the `tf.get_collection()` function." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "INFO:tensorflow:Restoring parameters from ./tf_logs/4_1_EQ/model\n" ] } ], "source": [ "tf.reset_default_graph()\n", "\n", "sess = tf.InteractiveSession()\n", "\n", "new_saver = tf.train.import_meta_graph(ckptdir + '.meta')\n", "new_saver.restore(sess, tf.train.latest_checkpoint(logdir))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 5. Attaching Subgraph for Calculating Relevance Scores" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "collapsed": true }, "outputs": [], "source": [ "conv_ksize = [1, 3, 3, 1]\n", "pool_ksize = [1, 2, 2, 1]\n", "conv_strides = [1, 1, 1, 1]\n", "pool_strides = [1, 2, 2, 1]\n", "\n", "activations = tf.get_collection('EC')\n", "weights = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES, scope='CNN')\n", "\n", "X = activations[0]\n", "predictions = activations[-1]\n", "\n", "activations.reverse()\n", "weights.reverse()\n", "\n", "taylor = Taylor(activations, weights, conv_ksize, pool_ksize, conv_strides, pool_strides, 'Taylor')\n", "\n", "Rs = [taylor(i) for i in range(10)]\n", "SA_scores = [tf.square(tf.gradients(predictions[:,i], X)) for i in range(10)]\n", "STD_scores = [tf.gradients(predictions[:,i], X) * X for i in range(10)]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### 6. Calculating Relevance Scores $R(x_i)$ and Displaying Images\n", "\n", "We can see that while Sensitivity Analysis and Simple Taylor Decomposition provides discontinuous explanations with frequent jumps, Deep Taylor Decomposition gives us a continuous explanation." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABDAAAAC1CAYAAACppg9uAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/w9tsRRFDmgMKAAAAABJRU5ErkJggg==\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABJ0AAAFVCAYAAABFM545AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n++3sWqkkq+MAAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "digit = 2\n", "dist = 41\n", "init = -20\n", "\n", "sample_imgs = [images[np.argmax(labels, axis=1) == i][3] for i in range(10)]\n", "translated_imgs = [[translate(img, init + i, 0) for i in range(dist)] for img in sample_imgs]\n", "\n", "preds = [[sess.run(predictions[:,i], feed_dict={X:translated_imgs[i][j]}) for j in range(dist)] for i in range(10)]\n", "\n", "sa_imgs = [[sess.run(SA_scores[i], feed_dict={X:translated_imgs[i][j]}) for j in range(dist)] for i in range(10)]\n", "sa_imgs = np.square(np.reshape(sa_imgs, [10, dist, -1]))\n", "\n", "std_imgs = [[sess.run(STD_scores[i], feed_dict={X:translated_imgs[i][j]}) for j in range(dist)] for i in range(10)]\n", "std_imgs = np.reshape(std_imgs, [10, dist, -1])\n", "\n", "dtd_imgs = [[sess.run(Rs[i], feed_dict={X:translated_imgs[i][j]}) for j in range(dist)] for i in range(10)]\n", "dtd_imgs = np.squeeze(dtd_imgs, axis=2)\n", "\n", "sa_imgs_q0 = sa_imgs[digit]\n", "sa_imgs_q1 = sa_imgs[digit].reshape(dist,28,28)[:,0:14,0:14].reshape(dist,-1)\n", "sa_imgs_q2 = sa_imgs[digit].reshape(dist,28,28)[:,0:14,14:28].reshape(dist,-1)\n", "sa_imgs_q3 = sa_imgs[digit].reshape(dist,28,28)[:,14:28,14:28].reshape(dist,-1)\n", "sa_imgs_q4 = sa_imgs[digit].reshape(dist,28,28)[:,14:28,0:14].reshape(dist,-1)\n", "\n", "std_imgs_q0 = std_imgs[digit]\n", "std_imgs_q1 = std_imgs[digit].reshape(dist,28,28)[:,0:14,0:14].reshape(dist,-1)\n", "std_imgs_q2 = std_imgs[digit].reshape(dist,28,28)[:,0:14,14:28].reshape(dist,-1)\n", "std_imgs_q3 = std_imgs[digit].reshape(dist,28,28)[:,14:28,14:28].reshape(dist,-1)\n", "std_imgs_q4 = std_imgs[digit].reshape(dist,28,28)[:,14:28,0:14].reshape(dist,-1)\n", "\n", "dtd_imgs_q0 = dtd_imgs[digit]\n", "dtd_imgs_q1 = dtd_imgs[digit].reshape(dist,28,28)[:,0:14,0:14].reshape(dist,-1)\n", "dtd_imgs_q2 = dtd_imgs[digit].reshape(dist,28,28)[:,0:14,14:28].reshape(dist,-1)\n", "dtd_imgs_q3 = dtd_imgs[digit].reshape(dist,28,28)[:,14:28,14:28].reshape(dist,-1)\n", "dtd_imgs_q4 = dtd_imgs[digit].reshape(dist,28,28)[:,14:28,0:14].reshape(dist,-1)\n", "\n", "fig = plt.figure(figsize=[15,15])\n", "for i in range(6):\n", " if i is 0:\n", " ax = fig.add_subplot(1, 6, 1)\n", " ax.imshow(np.zeros([28,28]), cmap='gray_r', vmin=0, vmax=1)\n", " ax.tick_params(labelbottom='off', labelleft='off', bottom='off', left='off')\n", " ax.axvline(14, linestyle='--', color='black')\n", " ax.axhline(14, linestyle='--', color='black')\n", " ax.text(2, 9, '$R_1$', fontdict={'fontsize': 50})\n", " ax.text(16, 9, '$R_2$', fontdict={'fontsize': 50})\n", " ax.text(16, 23, '$R_3$', fontdict={'fontsize': 50})\n", " ax.text(2, 23, '$R_4$', fontdict={'fontsize': 50})\n", " continue\n", " \n", " ax = fig.add_subplot(1, 6, i + 1)\n", " ax.imshow(dtd_imgs_q0[3 * i + 12].reshape([28,28]), cmap='hot_r')\n", " ax.tick_params(labelbottom='off', labelleft='off', bottom='off', left='off')\n", "plt.tight_layout()\n", "\n", "fig = plt.figure(figsize=[20,5])\n", "ax1 = fig.add_subplot(131)\n", "x = list(range(init,init+dist))\n", "y0 = preds[digit]\n", "y1 = np.sum(sa_imgs_q1, axis=1) / 2\n", "y2 = np.sum(sa_imgs_q2, axis=1) / 2\n", "y3 = np.sum(sa_imgs_q3, axis=1) / 2\n", "y4 = np.sum(sa_imgs_q4, axis=1) / 2\n", "\n", "ax1.plot(x, y0, marker='.', label='$f(x)$')\n", "ax1.plot(x, y1, marker='.', label='$\\propto R_1(x)$')\n", "ax1.plot(x, y2, marker='.', label='$\\propto R_2(x)$')\n", "ax1.plot(x, y3, marker='.', label='$\\propto R_3(x)$')\n", "ax1.plot(x, y4, marker='.', label='$\\propto R_4(x)$')\n", "ax1.legend(loc='upper left', fontsize='x-large')\n", "\n", "ax1.set_title('Sensitivity Analysis for Image Translation', fontdict={'fontsize': 14})\n", "ax1.set_xlabel('Digit Position', fontdict={'fontsize': 14})\n", "ax1.set_ylabel('Relevance Scores', fontdict={'fontsize': 14})\n", "ax1.tick_params(labelsize=12)\n", "\n", "ax2 = fig.add_subplot(132)\n", "y1 = np.sum(std_imgs_q1, axis=1)\n", "y2 = np.sum(std_imgs_q2, axis=1)\n", "y3 = np.sum(std_imgs_q3, axis=1)\n", "y4 = np.sum(std_imgs_q4, axis=1)\n", "\n", "ax2.plot(x, y0, marker='.', label='$f(x)$')\n", "ax2.plot(x, y1, marker='.', label='$R_1(x)$')\n", "ax2.plot(x, y2, marker='.', label='$R_2(x)$')\n", "ax2.plot(x, y3, marker='.', label='$R_3(x)$')\n", "ax2.plot(x, y4, marker='.', label='$R_4(x)$')\n", "ax2.legend(loc='upper left', fontsize='x-large')\n", "\n", "ax2.set_title('Simple Taylor Decomposition for Image Translation', fontdict={'fontsize': 14})\n", "ax2.set_xlabel('Digit Position', fontdict={'fontsize': 14})\n", "ax2.set_ylabel('Relevance Scores', fontdict={'fontsize': 14})\n", "ax2.tick_params(labelsize=12)\n", "\n", "ax3 = fig.add_subplot(133)\n", "y1 = np.sum(dtd_imgs_q1, axis=1)\n", "y2 = np.sum(dtd_imgs_q2, axis=1)\n", "y3 = np.sum(dtd_imgs_q3, axis=1)\n", "y4 = np.sum(dtd_imgs_q4, axis=1)\n", "\n", "ax3.plot(x, y0, marker='.', label='$f(x)$')\n", "ax3.plot(x, y1, marker='.', label='$R_1(x)$')\n", "ax3.plot(x, y2, marker='.', label='$R_2(x)$')\n", "ax3.plot(x, y3, marker='.', label='$R_3(x)$')\n", "ax3.plot(x, y4, marker='.', label='$R_4(x)$')\n", "ax3.legend(loc='upper left', fontsize='x-large')\n", "\n", "ax3.set_title('Deep Taylor Decomposition for Image Translation', fontdict={'fontsize': 14})\n", "ax3.set_xlabel('Digit Position', fontdict={'fontsize': 14})\n", "ax3.set_ylabel('Relevance Scores', fontdict={'fontsize': 14})\n", "ax3.tick_params(labelsize=12)" ] }, { "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": "3.6.2" }, "latex_envs": { "LaTeX_envs_menu_present": true, "autoclose": false, "autocomplete": true, "bibliofile": "biblio.bib", "cite_by": "apalike", "current_citInitial": 1, "eqLabelWithNumbers": true, "eqNumInitial": 1, "hotkeys": { "equation": "Ctrl-E", "itemize": "Ctrl-I" }, "labels_anchors": false, "latex_user_defs": false, "report_style_numbering": false, "user_envs_cfg": false } }, "nbformat": 4, "nbformat_minor": 2 }