Skip to content

Instantly share code, notes, and snippets.

@khalednasr
Last active August 29, 2015 14:04
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save khalednasr/b39fafd092854319869d to your computer and use it in GitHub Desktop.
Save khalednasr/b39fafd092854319869d to your computer and use it in GitHub Desktop.
{
"metadata": {
"name": "autoencoders"
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "heading",
"level": 1,
"metadata": {},
"source": [
"Deep Autoencoders"
]
},
{
"cell_type": "heading",
"level": 4,
"metadata": {},
"source": [
"by Khaled Nasr as a part of a <a href=\"https://www.google-melange.com/gsoc/project/details/google/gsoc2014/khalednasr92/5657382461898752\">GSoC 2014 project</a> mentored by Theofanis Karaletsos and Sergey Lisitsyn "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This notebook illustrates how to train and evaluate a deep autoencoder using Shogun"
]
},
{
"cell_type": "heading",
"level": 2,
"metadata": {},
"source": [
"Introduction"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"A (single layer) [autoencoder](http://deeplearning.net/tutorial/dA.html#autoencoders) is a neural network that has three layers: an input layer, a hidden (encoding) layer, and a decoding layer. The network is trained to reconstruct its inputs, which forces the hidden layer to try to learn good representations of the inputs.\n",
"\n",
"In order to encourage the hidden layer to learn good input representations, certain variations on the simple autoencoder exist. Shogun currently supports two of them: Denoising Autoencoders [1] and Contractive Autoencoders [2]. In this notebook we'll focus on denoising autoencoders. \n",
"\n",
"For denoising autoencoders, each time a new training example is introduced to the network, it's randomly corrupted in some mannar, and the target is set to the original example. The autoencoder will try to recover the orignal data from it's noisy version, which is why it's called a denoising autoencoder. This process will force the hidden layer to learn a good representation of the input, one which is not affected by the corruption process.\n",
"\n",
"A deep autoencoder is an autoencoder with multiple hidden layers. Training such autoencoders directly is usually difficult, however, they can be pre-trained as a stack of single layer autoencoders. That is, we train the first hidden layer to reconstruct the input data, and then train the second hidden layer to reconstruct the states of the first hidden layer, and so on. After pre-training, we can train the entire deep autoencoder to fine-tune all the parameters together. We can also use the autoencoder to initialize a regular neural network and train it in a supervised manner.\n",
"\n",
"In this notebook we'll apply deep autoencoders to the USPS dataset for handwritten digits. We'll start by loading the data and dividing it into a training set and a test set:"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from scipy.io import loadmat\n",
"from modshogun import RealFeatures, MulticlassLabels\n",
"\n",
"# load the dataset\n",
"dataset = loadmat('../../../data/multiclass/usps.mat')\n",
"\n",
"Xall = dataset['data']\n",
"# the usps dataset has the digits labeled from 1 to 10 \n",
"# we'll subtract 1 to make them in the 0-9 range instead\n",
"Yall = np.array(dataset['label'].squeeze(), dtype=np.double)-1 \n",
"\n",
"# 7000 examples for training\n",
"Xtrain = RealFeatures(Xall[:,0:7000])\n",
"Ytrain = MulticlassLabels(Yall[0:7000])\n",
"\n",
"# the rest for testing\n",
"Xtest = RealFeatures(Xall[:,7000:-1])\n",
"Ytest = MulticlassLabels(Yall[7000:-1])"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 12
},
{
"cell_type": "heading",
"level": 2,
"metadata": {},
"source": [
"Creating the autoencoder"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Similar to regular neural networks in Shogun, we create a [deep autoencoder](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CDeepAutoencoder.html) using an array of [NeuralLayer](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CNeuralLayer.html)-based classes, which can be created using the utility class [NeuralLayers](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CNeuralLayers.html). However, for deep autoencoders there's a restriction that the layer sizes in the network have to be symmetric, that is, the first layer has to have the same size as the last layer, the second layer has to have the same size as the second-to-last layer, and so on. This restriction is necessary for pre-training to work. More details on that can found in the following section.\n",
"\n",
"We'll create a 5-layer deep autoencoder with following layer sizes: 256->512->128->512->256. We'll use [rectified linear neurons](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CNeuralRectifiedLinearLayer.html) for the hidden layers and [linear neurons](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CNeuralLinearLayer.html) for the output layer."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from modshogun import NeuralLayers, DeepAutoencoder\n",
"\n",
"layers = NeuralLayers()\n",
"layers = layers.input(256).rectified_linear(512).rectified_linear(128).rectified_linear(512).linear(256).done()\n",
"\n",
"ae = DeepAutoencoder(layers)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 13
},
{
"cell_type": "heading",
"level": 2,
"metadata": {},
"source": [
"Pre-training"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we can pre-train the network. To illustrate exactly what's going to happen, we'll give the layers some labels: L1 for the input layer, L2 for the first hidden layer, and so on up to L5 for the output layer.\n",
"\n",
"In pre-training, an autoencoder will formed for each encoding layer (layers up to the middle layer in the network). So here we'll have two autoencoders: L1->L2->L5, and L2->L3->L4. The first autoencoder will be trained on the raw data and used to initialize the weights and biases of layers L2 and L5 in the deep autoencoder. After the first autoencoder is trained, we use it to transform the raw data into the states of L2. These states will then be used to train the second autoencoder, which will be used to initialize the weights and biases of layers L3 and L4 in the deep autoencoder.\n",
"\n",
"The operations described above are performed by the the [pre_train()](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CDeepAutoencoder.html#acf6896cb166afbba063fd1257cb8bc97) function. Pre-training parameters for each autoencoder can be controlled using the [pt_* public attributes](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CDeepAutoencoder.html#a6389a6f19b8854c64e1b6be5aa0c1fc4) of [DeepAutoencoder](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CDeepAutoencoder.html). Each of those attributes is an [SGVector](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1SGVector.html) whose length is the number of autoencoders in the deep autoencoder (2 in our case). It can be used to set the parameters for each autoencoder indiviually. [SGVector's set_const()](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1SGVector.html#a8bce01a1fc41a734d9b5cf1533fd7a2a) method can also be used to assign the same parameter value for all autoencoders.\n",
"\n",
"Different noise types can be used to corrupt the inputs in a denoising autoencoder. Shogun currently supports 2 [noise types](http://www.shogun-toolbox.org/doc/en/latest/namespaceshogun.html#af95cf5d3778127a87c8a67516405d863): dropout noise, where a random portion of the inputs is set to zero at each iteration in training, and gaussian noise, where the inputs are corrupted with random gaussian noise. The noise type and strength can be controlled using [pt_noise_type](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CDeepAutoencoder.html#af6e5d2ade5cb270cc50565d590f929ae) and [pt_noise_parameter](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CDeepAutoencoder.html#adbdff6c07fa7dd70aaf547e192365075). Here, we'll use dropout noise."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from modshogun import AENT_DROPOUT, NNOM_GRADIENT_DESCENT, MSG_INFO\n",
"\n",
"ae.pt_noise_type.set_const(AENT_DROPOUT) # use dropout noise\n",
"ae.pt_noise_parameter.set_const(0.5) # each input has a 50% chance of being set to zero\n",
"\n",
"ae.pt_optimization_method.set_const(NNOM_GRADIENT_DESCENT) # train using gradient descent\n",
"ae.pt_gd_learning_rate.set_const(0.01)\n",
"ae.pt_gd_mini_batch_size.set_const(128)\n",
"\n",
"ae.pt_max_num_epochs.set_const(100)\n",
"ae.pt_epsilon.set_const(0.0) # disable automatic convergence testing\n",
"\n",
"# allow the INFO messages to be printed to the console, useful for monitoring training progress\n",
"ae.io.set_loglevel(MSG_INFO) \n",
"\n",
"# start pre-training. this might take some time\n",
"ae.pre_train(Xtrain)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 14
},
{
"cell_type": "heading",
"level": 2,
"metadata": {},
"source": [
"Fine-tuning"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"After pre-training, we can train the autoencoder as a whole to fine-tune the parameters. Training the whole autoencoder is performed using the [train()](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CAutoencoder.html#ace3eb6cc545affcbfa31d754ffd087dc) function. Training parameters are controlled through the [public attributes](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CDeepAutoencoder.html#pub-attribs), same as a regular neural network."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"ae.noise_type = AENT_DROPOUT # same noise type we used for pre-training\n",
"ae.noise_parameter = 0.5\n",
"\n",
"ae.max_num_epochs = 100\n",
"ae.optimization_method = NNOM_GRADIENT_DESCENT\n",
"ae.gd_mini_batch_size = 128\n",
"ae.gd_learning_rate = 0.0001\n",
"ae.epsilon = 0.0\n",
"\n",
"# start fine-tuning. this might take some time\n",
"_ = ae.train(Xtrain)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 15
},
{
"cell_type": "heading",
"level": 2,
"metadata": {},
"source": [
"Evaluation"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we can evaluate the autoencoder that we trained. We'll start by providing it with corrupted inputs and looking at how it will reconstruct them. The function [reconstruct()](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CDeepAutoencoder.html#ae8c2d565cf2ea809103d0557c57689c7) is used to obtain the reconstructions:"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# get a 50-example subset of the test set\n",
"subset = Xtest[:,0:50].copy()\n",
"\n",
"# corrupt the first 25 examples with multiplicative noise\n",
"subset[:,0:25] *= (random.random((256,25))>0.5)\n",
"\n",
"# corrupt the other 25 examples with additive noise \n",
"subset[:,25:50] += random.random((256,25))\n",
"\n",
"# obtain the reconstructions\n",
"reconstructed_subset = ae.reconstruct(RealFeatures(subset))\n",
"\n",
"# plot the corrupted data and the reconstructions\n",
"figure(figsize=(10,10))\n",
"for i in range(50):\n",
" ax1=subplot(10,10,i*2+1)\n",
" ax1.imshow(subset[:,i].reshape((16,16)), interpolation='nearest', cmap = cm.Greys_r)\n",
" ax1.set_xticks([])\n",
" ax1.set_yticks([])\n",
"\n",
" ax2=subplot(10,10,i*2+2)\n",
" ax2.imshow(reconstructed_subset[:,i].reshape((16,16)), interpolation='nearest', cmap = cm.Greys_r)\n",
" ax2.set_xticks([])\n",
" ax2.set_yticks([])"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAjwAAAI8CAYAAAD1D3GaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n"
}
],
"prompt_number": 16
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The figure shows the corrupted examples and their reconstructions. The top half of the figure shows the ones corrupted with multiplicative noise, the bottom half shows the ones corrupted with additive noise. We can see that the autoencoders can provide decent reconstructions despite the heavy noise.\n",
"\n",
"Next we'll look at the weights that the first hidden layer has learned. To obtain the weights, we can call the [get_layer_parameters()]() function, which will return a vector containing both the weights and the biases of the layer. The biases are stored first in the array followed by the weights matrix in column-major format."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# obtain the weights matrix of the first hidden layer\n",
"# the 512 is the number of biases in the layer (512 neurons)\n",
"# the transpose is because numpy stores matrices in row-major format, and Shogun stores \n",
"# them in column major format\n",
"w1 = ae.get_layer_parameters(1)[512:].reshape(256,512).T\n",
"\n",
"# visualize the weights between the first 100 neurons in the hidden layer \n",
"# and the neurons in the input layer\n",
"figure(figsize=(10,10))\n",
"for i in range(100):\n",
"\tax1=subplot(10,10,i+1)\n",
"\tax1.imshow(w1[i,:].reshape((16,16)), interpolation='nearest', cmap = cm.Greys_r)\n",
"\tax1.set_xticks([])\n",
"\tax1.set_yticks([])"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAjwAAAI8CAYAAAD1D3GaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n"
}
],
"prompt_number": 17
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now, we can use the autoencoder to initialize a supervised neural network. The network will have all the layer of the autoencoder up to (and including) the middle layer. We'll also add a softmax output layer. So, the network will look like: L1->L2->L3->Softmax. The network is obtained by calling [convert_to_neural_network()](http://www.shogun-toolbox.org/doc/en/latest/classshogun_1_1CDeepAutoencoder.html#a8c179cd9a503b2fa78b9bfe10ae473e5):"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from modshogun import NeuralSoftmaxLayer\n",
"\n",
"nn = ae.convert_to_neural_network(NeuralSoftmaxLayer(10))\n",
"\n",
"nn.max_num_epochs = 50\n",
"\n",
"nn.io.set_loglevel(MSG_INFO)\n",
"\n",
"nn.set_labels(Ytrain)\n",
"_ = nn.train(Xtrain)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 18
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Next, we'll evaluate the accuracy on the test set:"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from modshogun import MulticlassAccuracy\n",
"\n",
"predictions = nn.apply_multiclass(Xtest)\n",
"accuracy = MulticlassAccuracy().evaluate(predictions, Ytest) * 100\n",
"\n",
"print \"Classification accuracy on the test set =\", accuracy, \"%\""
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Classification accuracy on the test set = 94.6451893774 %\n"
]
}
],
"prompt_number": 19
},
{
"cell_type": "heading",
"level": 2,
"metadata": {},
"source": [
"References"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"- [1] [Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion, Vincent, 2010](http://jmlr.org/papers/volume11/vincent10a/vincent10a.pdf)\n",
"- [2] [Contractive Auto-Encoders: Explicit Invariance During Feature Extraction, Rifai, 2011](http://machinelearning.wustl.edu/mlpapers/paper_files/ICML2011Rifai_455.pdf)"
]
}
],
"metadata": {}
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment