{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "It's been 6 months since my last blog. There are multiple blog drafts to be honest, but many of them took a really long time to finish. I guess I should have sliced the material a bit to multiple blogs. Anyway in this blog, I want to show how to achieve more than 95% accuracy with just Macbook Air. So let's dig down to details.\n", "\n", "This material originally comes [Deep Learning Lecture 4 from Udacity](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/udacity/4_convolutions.ipynb). In this blog, Pickling, reformat, accuracy, and session are theirs, but the architecture is my own which is the core of Deep Learning. It's kind of refreshing because I have experience it before (yes, Andrew Ng's Coursera Machine Learning on Neural Network).\n", "" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "cellView": "both", "colab": { "autoexec": { "startup": false, "wait_interval": 0 } }, "colab_type": "code", "collapsed": true, "id": "tm2CQN_Cpwj0" }, "outputs": [], "source": [ "from __future__ import print_function\n", "import numpy as np\n", "import tensorflow as tf\n", "from six.moves import cPickle as pickle\n", "from six.moves import range\n", "\n", "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", "import string" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The original MNIST dataset has been pickled and reshaped into training, validation and test set at [Lecture 1](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/udacity/1_notmnist.ipynb). We have dataset with only A-J letters to predict (limited because of computing power)." ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "cellView": "both", "colab": { "autoexec": { "startup": false, "wait_interval": 0 }, "output_extras": [ { "item_id": 1 } ] }, "colab_type": "code", "collapsed": false, "executionInfo": { "elapsed": 11948, "status": "ok", "timestamp": 1446658914837, "user": { "color": "", "displayName": "", "isAnonymous": false, "isMe": true, "permissionId": "", "photoUrl": "", "sessionId": "0", "userId": "" }, "user_tz": 480 }, "id": "y3-cj1bpmuxc", "outputId": "016b1a51-0290-4b08-efdb-8c95ffc3cd01" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training set (200000, 28, 28) (200000,)\n", "Validation set (10000, 28, 28) (10000,)\n", "Test set (10000, 28, 28) (10000,)\n" ] } ], "source": [ "pickle_file = 'notMNIST.pickle'\n", "\n", "with open(pickle_file, 'rb') as f:\n", " save = pickle.load(f)\n", " train_dataset = save['train_dataset']\n", " train_labels = save['train_labels']\n", " valid_dataset = save['valid_dataset']\n", " valid_labels = save['valid_labels']\n", " test_dataset = save['test_dataset']\n", " test_labels = save['test_labels']\n", " del save # hint to help gc free up memory\n", " print('Training set', train_dataset.shape, train_labels.shape)\n", " print('Validation set', valid_dataset.shape, valid_labels.shape)\n", " print('Test set', test_dataset.shape, test_labels.shape)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's plot the first three images in dataset." ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "collapsed": true }, "outputs": [], "source": [ "def show_label_image(idx):\n", " print('Labels: ', string.ascii_uppercase[train_labels[idx]])\n", " return plt.imshow(train_dataset[idx])" ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Labels: E\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV8AAAFfCAYAAADptc+BAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "show_label_image(0);" ] }, { "cell_type": "code", "execution_count": 33, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Labels: J\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV8AAAFfCAYAAADptc+BAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAAAAAAAAAAAAAP52Aeogv2T5Sc1OAAAAAElFTkSuQmCC\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "show_label_image(1);" ] }, { "cell_type": "code", "execution_count": 35, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Labels: G\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV8AAAFfCAYAAADptc+BAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nWPC48f8BsIAneqV8aH4AAAAASUVORK5CYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "show_label_image(2);" ] }, { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "L7aHrm6nGDMB" }, "source": [ "Next, Reformat into a TensorFlow-friendly shape:\n", "- convolutions need the image data formatted as a cube (width by height by #channels)\n", "- labels as float 1-hot encodings." ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "cellView": "both", "colab": { "autoexec": { "startup": false, "wait_interval": 0 }, "output_extras": [ { "item_id": 1 } ] }, "colab_type": "code", "collapsed": false, "executionInfo": { "elapsed": 11952, "status": "ok", "timestamp": 1446658914857, "user": { "color": "", "displayName": "", "isAnonymous": false, "isMe": true, "permissionId": "", "photoUrl": "", "sessionId": "0", "userId": "" }, "user_tz": 480 }, "id": "IRSyYiIIGIzS", "outputId": "650a208c-8359-4852-f4f5-8bf10e80ef6c" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training set (200000, 28, 28, 1) (200000, 10)\n", "Validation set (10000, 28, 28, 1) (10000, 10)\n", "Test set (10000, 28, 28, 1) (10000, 10)\n" ] } ], "source": [ "image_size = 28\n", "num_labels = 10\n", "num_channels = 1 # grayscale\n", "\n", "import numpy as np\n", "\n", "def reformat(dataset, labels):\n", " dataset = dataset.reshape(\n", " (-1, image_size, image_size, num_channels)).astype(np.float32)\n", " labels = (np.arange(num_labels) == labels[:,None]).astype(np.float32)\n", " return dataset, labels\n", "train_dataset, train_labels = reformat(train_dataset, train_labels)\n", "valid_dataset, valid_labels = reformat(valid_dataset, valid_labels)\n", "test_dataset, test_labels = reformat(test_dataset, test_labels)\n", "print('Training set', train_dataset.shape, train_labels.shape)\n", "print('Validation set', valid_dataset.shape, valid_labels.shape)\n", "print('Test set', test_dataset.shape, test_labels.shape)" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "cellView": "both", "colab": { "autoexec": { "startup": false, "wait_interval": 0 } }, "colab_type": "code", "collapsed": true, "id": "AgQDIREv02p1" }, "outputs": [], "source": [ "def accuracy(predictions, labels):\n", " return (100.0 * np.sum(np.argmax(predictions, 1) == np.argmax(labels, 1))\n", " / predictions.shape[0])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "There is two major section when running Tensorflow. The first one is how you describe the deep learning architecture in form of tensor graph. Since we only describe and compile the architecture, there's no computation going on. The second one is session. It's made to tell the graph to actually play and learning from the data. We can use the session to report the validation set. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Okay, so this is where I play around over some parameters and deep learning architecture, which is where the real fun begin." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [ "batch_size = 128\n", "regularization_c = 0.12\n", "learning_rate_c = 0.4\n", "num_steps = 15001" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On machine learning side, there are these usual parameters that you have to tuned. Batch size tells you how many training set for the algorithm to tell at one step. Batching this lead you to stochastic gradient descent. Using smaller batch size makes the machine train faster, but less convergence to the true features that can predict the data. So you have to increase the number of steps to mitigate the issue. On the other hand, if you have bigger batch size, the machine train longer at each step, but converge at a less stochastic movement. So you have a smaller number of steps to train. I choose 128 images per step and 15000 number of steps.\n", "\n", "And then there's also some regularization parameter that let machine penalize features that overmagnified (prevent overfitting). Learning rates also tells how faster you want machine learning to train. I choose 0.4 and decay exponentially until 300 steps." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [ "dropout_rate = 0.5\n", "def get_relu(logits):\n", " relu = tf.nn.relu(logits)\n", " if train:\n", " return tf.nn.dropout(relu, dropout_rate)\n", " else:\n", " return relu" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On the deep learning side, I also set dropout rate of 0.5. That means if one layer in neural network contains 100 activation units, at each step 50% of those units will be selected randomly and perform forward prop and backprop. Intuitively, the layer can't rely on particular activation unit to make a prediction. Like regularization, dropout is used to prevent overfitting." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Don't forget to disable dropout at validation and test set. Because you want 100% (instead of 50%) prediction power for your deep learning." ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "collapsed": true }, "outputs": [], "source": [ "patch_size = 13\n", "depth = 16" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now here is where we get the core parameters of deep learning. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Convolutions and pooling in Tensorflow typically expect 4 inputs; data, kernel size, stride movement, and padding. There are additional weights (machine learning features) for convolutions. \n", "\n", "There is 2D patch(kernel) that stride over our images and summarize it by making smaller number of pixels but longer depth. This is what's called 2D Convolutions. Here you can see at first layer of 2D convolutions I make 13x13 2D patch and depth 16. That means, 28x28 images of grayscale (1x channel, instead of RGB 3x channels) will be made smaller to 16x16 but longer to 16 depth channels. So imagine 16 2D matrices of 16x16." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [ "pooling_size = 2" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Another method called pooling used to summarize the 2D matrix to even smaller number of pixels. The advantage of pooling over convolutions is that machine doesn't require additional weights to tuned. The pooling, like it called, creates a pooling at every stride movement of the matrix, and determine which value in pixels get in, and who doesn't. Max pooling then choose whichever pixel value the most in 2D matrix. Another method is average pooling, where you average all values in the matrix and pass in result to the next stage. Note that pooling doesn't increase the depths." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [ "stride_movement = [1, 2, 2, 1]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We keep discussing stride movement for convolutions and pooling. It requires 4 shape value; stride movement over data, width, height, and depth.\n", "\n", "`[1,2,2,1]` tells Tensorflow stride every data, stride every 2 width and height pixels, and stride every depth channels. This makes at every additional convolutions and pooling, 2D pixels gets halved." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "There are two different kinds of padding, same padding and valid padding. In valid padding the corner of the kernel started at the corner of the image. This makes the center of the kernel started and finished not in the edge of the images. In same padding you add zeros padding in edges of your 2D matrices, so the center of the kernels started at the corner of 2D images. \n", "\n", "Different padding, kernel size and stride movement will determine the size of output matrices." ] }, { "cell_type": "code", "execution_count": null, "metadata": { "collapsed": true }, "outputs": [], "source": [ "num_hidden = 64\n", "def get_weights(shape):\n", " return tf.Variable(tf.truncated_normal(shape, stddev=0.1))\n", "\n", "\n", "def get_biases(shape):\n", " return tf.Variable(tf.constant(1.0, shape=[shape[-1]]))\n", "\n", "\n", "def get_weights_biases(shape):\n", " weights = get_weights(shape)\n", " biases = get_biases(shape)\n", " return weights, biases" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Finally, I created convenient functions to get weights and biases. So you can just input the shape of the matrix you want, and return tensor variable. There is 4 shape value for 2D convolutions; width, height, input channels, and output channels. Pooling doesn't requires weight and bias. And there are only two shape value for the fully connected layer and logits; the number of activation input and number of activation output. For the fully connected layer, I choose 64 number of activation units as these are the final number of depth channels. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Here are layers of my deep learning architecture:\n", "\n", "1. 2D Convolutions (k=13, s=1, padding=valid)\n", "2. Max Pooling (k=2, s=2, padding=same)\n", "3. 2D Convolution (k=2, s=2, padding=same)\n", "4. Max Pooling (k=2, s=2, padding=same)\n", "5. 1x1 Convolutions (k=1, s=1, padding=valid)\n", "6. Fully Connected\n", "7. Fully Connected \n", "8. Classifier" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As we discussed with each of the layer, 2D image pixels gets halved at every additional layer." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "```\n", "28x28 (2D Conv) \n", " --> 16x16 (Max Pooling) \n", " --> 8x8 (2D Conv) \n", " --> 4x4 (Max Pooling) \n", " --> 2x2 (1x1 Conv) \n", " --> 1x1 Fully connected\n", " --> 1x1 Fully connected\n", " --> Classifier\n", "```" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In return, the depths get increased at every additional convolution layer." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "```\n", "1 grayscale channel (2D Conv) \n", " --> 16 (Max Pooling) \n", " --> 16 (2D Conv) \n", " --> 32 (Max Pooling) \n", " --> 32 (1x1 Conv) \n", " --> 64 (Fully Connected) \n", " --> 64 (Fully Connected) \n", " --> Classifier\n", "```" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is why convolutions talk about pixels get summarized with an increased depth. Hopefully all of the necessary details explained, and this is what the actual tensor graph looks like." ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "collapsed": false }, "outputs": [], "source": [ "graph = tf.Graph()\n", "\n", "with graph.as_default():\n", "\n", " # Input data.\n", " tf_train_dataset = tf.placeholder(\n", " tf.float32, shape=(batch_size, image_size, image_size, num_channels))\n", " tf_train_labels = tf.placeholder(tf.float32, shape=(batch_size, num_labels))\n", " tf_valid_dataset = tf.constant(valid_dataset)\n", " tf_test_dataset = tf.constant(test_dataset)\n", " \n", " # Variables.\n", " layer1_weights = get_weights([patch_size, patch_size, num_channels, depth])\n", " layer1_biases = tf.Variable(tf.zeros([depth]))\n", "\n", " layer3_weights, layer3_biases = get_weights_biases([4, 4, depth, depth*2])\n", " layer5_weights, layer5_biases = get_weights_biases([2, 2, depth*2, num_hidden])\n", " layer6_weights, layer6_biases = get_weights_biases([num_hidden, num_hidden])\n", " layer7_weights, layer7_biases = get_weights_biases([num_hidden, num_hidden])\n", " layer8_weights, layer8_biases = get_weights_biases([num_hidden, num_labels])\n", " \n", " # Model.\n", " def model(data, train=False):\n", " \n", " conv = tf.nn.conv2d(data, layer1_weights, [1, 1, 1, 1], padding='VALID')\n", " hidden = tf.nn.relu(conv + layer1_biases)\n", " \n", " pool = tf.nn.max_pool(hidden, [1, pooling_size, pooling_size, 1], stride_movement, padding='SAME')\n", " hidden = tf.nn.relu(pool)\n", " \n", " conv = tf.nn.conv2d(hidden, layer3_weights, stride_movement, padding='SAME')\n", " hidden = tf.nn.relu(conv + layer3_biases)\n", " \n", " pool = tf.nn.max_pool(hidden, [1, pooling_size, pooling_size, 1], stride_movement, padding='SAME')\n", " hidden = tf.nn.relu(pool)\n", " \n", " conv = tf.nn.conv2d(hidden, layer5_weights, [1, 1, 1, 1], padding='VALID')\n", " hidden = tf.nn.relu(conv + layer5_biases)\n", " \n", " shape = hidden.get_shape().as_list()\n", " reshape = tf.reshape(hidden, [shape[0], shape[1] * shape[2] * shape[3]])\n", " \n", " def get_relu(logits):\n", " relu = tf.nn.relu(logits)\n", " if train:\n", " return tf.nn.dropout(relu, dropout_rate)\n", " else:\n", " return relu\n", "\n", " fully_connected = tf.matmul(reshape, layer6_weights) + layer6_biases\n", " hidden = get_relu(fully_connected)\n", " \n", " fully_connected = tf.matmul(hidden, layer7_weights) + layer7_biases\n", " hidden = get_relu(fully_connected)\n", " \n", " classifier = tf.matmul(hidden, layer8_weights) + layer8_biases\n", " \n", " return classifier\n", " \n", " # Training computation.\n", " logits = model(tf_train_dataset, train=True)\n", " loss = tf.reduce_mean(\n", " tf.nn.softmax_cross_entropy_with_logits(logits, tf_train_labels))\n", " \n", " l2_loss = loss + (regularization_c * tf.nn.l2_loss(layer8_weights))\n", " global_step = tf.Variable(0)\n", " learning_rate = tf.train.exponential_decay(learning_rate_c, global_step, 300, 0.99 )\n", " \n", " # Optimizer.\n", " optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(l2_loss, global_step=global_step)\n", " \n", " # Predictions for the training, validation, and test data.\n", " train_prediction = tf.nn.softmax(logits)\n", " valid_prediction = tf.nn.softmax(model(tf_valid_dataset))\n", " test_prediction = tf.nn.softmax(model(tf_test_dataset))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now let's create a session to run the tensor graphs. You initialized the tensor variable before you run. Again, you can see the minibatch take turns at every step with offset. There is also `feed_dict` variable that let the data and labels injected at each session run. And we reported training score and validation score at each of the step, and test score at final step. Here the session runs and the reported accuracy." ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "collapsed": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Initialized\n", "Minibatch loss at step 0: 2.763546\n", "Minibatch accuracy: 10.2%\n", "Validation accuracy: 10.0%\n", "Minibatch loss at step 1000: 0.664659\n", "Minibatch accuracy: 82.8%\n", "Validation accuracy: 85.0%\n", "Minibatch loss at step 2000: 0.428772\n", "Minibatch accuracy: 87.5%\n", "Validation accuracy: 86.2%\n", "Minibatch loss at step 3000: 0.602961\n", "Minibatch accuracy: 82.8%\n", "Validation accuracy: 86.9%\n", "Minibatch loss at step 4000: 0.684882\n", "Minibatch accuracy: 77.3%\n", "Validation accuracy: 77.7%\n", "Minibatch loss at step 5000: 0.353280\n", "Minibatch accuracy: 89.8%\n", "Validation accuracy: 87.9%\n", "Minibatch loss at step 6000: 0.346389\n", "Minibatch accuracy: 89.1%\n", "Validation accuracy: 88.5%\n", "Minibatch loss at step 7000: 0.386851\n", "Minibatch accuracy: 87.5%\n", "Validation accuracy: 88.4%\n", "Minibatch loss at step 8000: 0.351568\n", "Minibatch accuracy: 89.1%\n", "Validation accuracy: 88.8%\n", "Minibatch loss at step 9000: 0.314128\n", "Minibatch accuracy: 89.8%\n", "Validation accuracy: 89.1%\n", "Minibatch loss at step 10000: 0.338578\n", "Minibatch accuracy: 88.3%\n", "Validation accuracy: 89.1%\n", "Minibatch loss at step 11000: 0.228507\n", "Minibatch accuracy: 93.0%\n", "Validation accuracy: 89.8%\n", "Minibatch loss at step 12000: 0.228184\n", "Minibatch accuracy: 93.0%\n", "Validation accuracy: 89.6%\n", "Minibatch loss at step 13000: 0.268132\n", "Minibatch accuracy: 93.0%\n", "Validation accuracy: 89.5%\n", "Minibatch loss at step 14000: 0.324846\n", "Minibatch accuracy: 89.8%\n", "Validation accuracy: 90.1%\n", "Minibatch loss at step 15000: 0.283048\n", "Minibatch accuracy: 89.8%\n", "Validation accuracy: 90.0%\n", "Test accuracy: 95.4%\n" ] } ], "source": [ "with tf.Session(graph=graph) as session:\n", " tf.global_variables_initializer().run()\n", " print('Initialized')\n", " for step in range(num_steps):\n", " offset = (step * batch_size) % (train_labels.shape[0] - batch_size)\n", " batch_data = train_dataset[offset:(offset + batch_size), :, :, :]\n", " batch_labels = train_labels[offset:(offset + batch_size), :]\n", " feed_dict = {tf_train_dataset : batch_data, tf_train_labels : batch_labels}\n", " _, l, predictions = session.run(\n", " [optimizer, loss, train_prediction], feed_dict=feed_dict)\n", " if (step % 1000 == 0):\n", " print('Minibatch loss at step %d: %f' % (step, l))\n", " print('Minibatch accuracy: %.1f%%' % accuracy(predictions, batch_labels))\n", " print('Validation accuracy: %.1f%%' % accuracy(\n", " valid_prediction.eval(), valid_labels))\n", " print('Test accuracy: %.1f%%' % accuracy(test_prediction.eval(), test_labels))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "You can see that even at every 1000 steps, the validation can increase/decrease. This is the result of the mini batch gradient descent that perform stochastic movement to converge ultimate predictive power. Let's do some of the back of the envelope calculations.\n", "\n", "We have 128 images and 15000 of steps to run. The number of images that the machine saw, " ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "1920000" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "128 * 15000" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "While we have 200,000 images, the machine at least saw one image," ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "9.6" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "1920000. / 200000" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "So the machine saw almost 10 times at each image. This is the same as running full dataset on just 10 steps. This is relatively small number, and so the machine has potential room to grow by just increasing number of steps, but resulting to longer time to compute." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Postscript" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Obviously, there are some missing details in this blog. Namely,\n", "\n", "* End-to-end deep learning process\n", "* Computation behind convolutions\n", "* Neural Network definition\n", "* Arithmetic to get required output number of pixels based on input size, kernel size, stride, padding\n", "* Machine Learning classification of one-hot-encoding\n", "* Softmax and Cross Entropy\n", "* Non-linear function like relu(used in this blog), sigmoid and tanh." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "While I would love talk about it, I prefer to keep this blog short. That material alone could be made for another blog. Here I just want to share my deep learning structure and reported accuracy. This is relatively simple deep learning architecture. There are hundreds of layer in deep learning architecture in major production to infer complex interaction in an image rather than simple letter recognition. And there is another thing called inception, that use various combination of pooling and convolutions in one layer. All of these require multiple GPU units for deep learning to train faster.\n", "\n", "We also talked about pooling, convolutions, and fully connected. Hopefully you understand if you see other CNN deep learning architecture in the future. \n", "\n", "You can download this notebook and try it yourself. See if you can play around with the parameters and beat the accuracy in this blog. And if this blog still feels like missing some details to you, I encourage you to take [Deep Learning course by Google at Udacity](https://www.udacity.com/course/deep-learning--ud730). It has all the information you need and more." ] } ], "metadata": { "colab": { "default_view": {}, "name": "4_convolutions.ipynb", "provenance": [], "version": "0.3.2", "views": {} }, "kernelspec": { "display_name": "Python 2", "language": "python", "name": "python2" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 2 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython2", "version": "2.7.10" } }, "nbformat": 4, "nbformat_minor": 0 }