Exercise: Implement deep networks for digit classification

From Ufldl

Jump to: navigation, search
 
Line 33: Line 33:
We first forward propagate the training set through the first autoencoder (using <tt>feedForwardAutoencoder.m</tt> that you completed in [[Exercise:Self-Taught_Learning]]) to obtain hidden unit activations. These activations are then used to train the second sparse autoencoder. Since this is just an adapted application of a standard autoencoder, it should run similarly with the first. Complete this part of the code so as to learn a first layer of features using your <tt>sparseAutoencoderCost.m</tt> and minFunc.
We first forward propagate the training set through the first autoencoder (using <tt>feedForwardAutoencoder.m</tt> that you completed in [[Exercise:Self-Taught_Learning]]) to obtain hidden unit activations. These activations are then used to train the second sparse autoencoder. Since this is just an adapted application of a standard autoencoder, it should run similarly with the first. Complete this part of the code so as to learn a first layer of features using your <tt>sparseAutoencoderCost.m</tt> and minFunc.
 +
 +
This part of the exercise demonstrates the idea of greedy layerwise training with the ''same'' learning algorithm reapplied multiple times.
=== Step 3: Train the softmax classifier on the L2 features ===
=== Step 3: Train the softmax classifier on the L2 features ===
-
Next, continue to forward propagate the L1 features through the second autoencoder (using <tt>feedForwardAutoencoder.m</tt>) to obtain the L2 hidden unit activations. These activations are then used to train the softmax classifier. You should be able to use <tt>softmaxTrain.m</tt> that you completed in [[Exercise:Softmax Regression]] to complete this part of the assignment.
+
Next, continue to forward propagate the L1 features through the second autoencoder (using <tt>feedForwardAutoencoder.m</tt>) to obtain the L2 hidden unit activations. These activations are then used to train the softmax classifier. You can either use <tt>softmaxTrain.m</tt> or directly use <tt>softmaxCost.m</tt> that you completed in [[Exercise:Softmax Regression]] to complete this part of the assignment.
=== Step 4: Implement fine-tuning ===
=== Step 4: Implement fine-tuning ===
-
To implement fine tuning, we need to consider all three layers as a single model. Implement <tt>stackedAECost.m</tt> to return the cost and gradient of the model. The cost function should be as defined as the log likelihood and a gradient decay term. The gradient should be computed using [[Backpropagation Algorithm | back-propogation as discussed earlier]]. The predictions should consist of the activations of the output layer of the softmax model.
+
To implement fine tuning, we need to consider all three layers as a single model. Implement <tt>stackedAECost.m</tt> to return the cost and gradient of the model. The cost function should be as defined as the log likelihood and a gradient decay term. The gradient should be computed using [[Backpropagation Algorithm | back-propagation as discussed earlier]]. The predictions should consist of the activations of the output layer of the softmax model.
To help you check that your implementation is correct, you should also check your gradients on a synthetic small dataset. We have implemented <tt>checkStackedAECost.m</tt> to help you check your gradients. If this checks passes, you will have implemented fine-tuning correctly.
To help you check that your implementation is correct, you should also check your gradients on a synthetic small dataset. We have implemented <tt>checkStackedAECost.m</tt> to help you check your gradients. If this checks passes, you will have implemented fine-tuning correctly.
-
{{Quote|
+
'''Note:''' When adding the weight decay term to the cost, you should regularize only the softmax weights (do not regularize the weights that compute the hidden layer activations).
-
Note: When adding the weight decay term to the cost, you should regularize '''all''' the weights in the network.
+
 
-
}}
+
'''Implementation Tip:''' It is always a good idea to implement the code modularly and check (the gradient of) each part of the code before writing the more complicated parts.
=== Step 5: Test the model ===
=== Step 5: Test the model ===
Line 52: Line 54:
Finally, you will need to classify with this model; complete the code in <tt>stackedAEPredict.m</tt> to classify using the stacked autoencoder with a classification layer.
Finally, you will need to classify with this model; complete the code in <tt>stackedAEPredict.m</tt> to classify using the stacked autoencoder with a classification layer.
-
After completing these steps, running the entire script in stackedAETrain.m will perform layer-wise training of the stacked autoencoder, finetune the model, and measure its performance on the test set. If you've done all the steps correctly, you should get an accuracy of about XX.X percent.
+
After completing these steps, running the entire script in stackedAETrain.m will perform layer-wise training of the stacked autoencoder, finetune the model, and measure its performance on the test set. If you've done all the steps correctly, you should get an accuracy of about 87.7% before finetuning and 97.6% after finetuning (for the 10-way classification problem).
 +
 
 +
 
 +
{{CNN}}

Latest revision as of 11:04, 26 May 2011

Personal tools