Home work 3: Basic Artificial Neural Networks

The goal of this homework is simple, yet an actual implementation may take some time :). We are going to write an Artificial Neural Network (almost) from scratch. The software design of was heavily inspired by Torch which is the most convenient neural network environment when the work involves defining new layers.

This homework requires sending "multiple files, please do not forget to include all the files when sending to TA. The list of files:

  • This notebook
  • HW3_Modules.ipynb
  • HW3_differentiation.ipynb

In [ ]:
%matplotlib inline
from time import time, sleep
import numpy as np
import matplotlib.pyplot as plt
from IPython import display

Framework

Implement everything in Modules.ipynb. Read all the comments thoughtfully to ease the pain. Please try not to change the prototypes.

Do not forget, that each module should return AND store output and gradInput.

The typical assumption is that module.backward is always executed after module.forward, so output is stored, this would be useful for SoftMax.


In [ ]:
"""
    --------------------------------------
    -- Tech note
    --------------------------------------
    Inspired by torch I would use
    
    np.multiply, np.add, np.divide, np.subtract instead of *,+,/,-
    for better memory handling
        
    Suppose you allocated a variable    
        
        a = np.zeros(...)
    
    So, instead of
    
        a = b + c  # will be reallocated, GC needed to free
    
    I would go for: 
    
        np.add(b,c,out = a) # puts result in `a`
    
    But it is completely up to you.
"""
%run HW3_Modules.ipynb

Optimizer is implemented for you.


In [ ]:
def sgd_momentum(x, dx, config, state):
    """
        This is a very ugly implementation of sgd with momentum 
        just to show an example how to store old grad in state.
        
        config:
            - momentum
            - learning_rate
        state:
            - old_grad
    """
    
    # x and dx have complex structure, old dx will be stored in a simpler one
    state.setdefault('old_grad', {})
    
    i = 0 
    for cur_layer_x, cur_layer_dx in zip(x,dx): 
        for cur_x, cur_dx in zip(cur_layer_x,cur_layer_dx):
            
            cur_old_grad = state['old_grad'].setdefault(i, np.zeros_like(cur_dx))
            
            np.add(config['momentum'] * cur_old_grad, config['learning_rate'] * cur_dx, out = cur_old_grad)
            
            cur_x -= cur_old_grad
            i += 1

Toy example

Use this example to debug your code, start with logistic regression and then test other layers. You do not need to change anything here. This code is provided for you to test the layers. Also it is easy to use this code in MNIST task.


In [ ]:
# Generate some data
N = 500

X1 = np.random.randn(N,2) + np.array([2,2])
X2 = np.random.randn(N,2) + np.array([-2,-2])

Y = np.concatenate([np.ones(N),np.zeros(N)])[:,None]
Y = np.hstack([Y, 1-Y])

X = np.vstack([X1,X2])
plt.scatter(X[:,0],X[:,1], c = Y[:,0], edgecolors= 'none')

Define a logistic regression for debugging.


In [ ]:
net = Sequential()
net.add(Linear(2, 2))
net.add(SoftMax())

criterion = ClassNLLCriterion()

print net

# Test something like that then 

# net = Sequential()
# net.add(Linear(2, 4))
# net.add(ReLU())
# net.add(Linear(4, 2))
# net.add(SoftMax())

Start with batch_size = 1000 to make sure every step lowers the loss, then try stochastic version.


In [ ]:
# Iptimizer params
optimizer_config = {'learning_rate' : 1e-1, 'momentum': 0.9}
optimizer_state = {}

# Looping params
n_epoch = 20
batch_size = 128

In [ ]:
# batch generator
def get_batches( (X, Y) , batch_size):
    n_samples = X.shape[0]
        
    # Shuffle at the start of epoch
    indices = np.arange(n_samples)
    np.random.shuffle(indices)
    
    for start in range(0, n_samples, batch_size):
        end = min(start + batch_size, n_samples)
        
        batch_idx = indices[start:end]
    
        yield X[batch_idx], Y[batch_idx]

Train

Basic training loop. Examine it.


In [ ]:
loss_history = []

for i in range(n_epoch):
    for x_batch, y_batch in get_batches( (X,Y) , batch_size):
        
        net.zeroGradParameters()
        
        # Forward
        predictions = net.forward(x_batch)
        loss = criterion.forward(predictions, y_batch)
    
        # Backward
        dp = criterion.backward(predictions, y_batch)
        net.backward(x_batch, dp)
        
        # Update weights
        sgd_momentum(net.getParameters(), 
                     net.getGradParameters(), 
                     optimizer_config,
                     optimizer_state)      
        
        loss_history.append(loss)

    # Visualize
    display.clear_output(wait=True)
    plt.figure(figsize=(8, 6))
        
    plt.title("Training loss")
    plt.xlabel("#iteration")
    plt.ylabel("loss")
    plt.plot(loss_history, 'b')
    plt.show()
    
    print('Current loss: %f' % loss)

Digit classification

We are using MNIST as our dataset. Lets start with cool visualization. The most beautiful demo is the second one, if you are not familiar with convolutions you can return to it in several lectures.


In [ ]:
import os
from sklearn.datasets import fetch_mldata

# Fetch MNIST dataset and create a local copy.
if os.path.exists('mnist.npz'):
    with np.load('mnist.npz', 'r') as data:
        X = data['X']
        y = data['y']
else:
    mnist = fetch_mldata("mnist-original")
    X, y = mnist.data / 255.0, mnist.target
    np.savez('mnist.npz', X=X, y=y)

One-hot encode the labels first.


In [ ]:
# Your code goes here. ################################################
  • Compare ReLU, ELU, LeakyReLU, SoftPlus activation functions. You would better pick the best optimizer params for each of them, but it is overkill for now. Use an architecture of your choice for the comparison.
  • Try inserting BatchMeanSubtraction between Linear module and activation functions.
  • Plot the losses both from activation functions comparison and BatchMeanSubtraction comparison on one plot. Please find a scale (log?) when the lines are distinguishable, do not forget about naming the axes, the plot should be goodlooking.
  • Hint: logloss for MNIST should be around 0.5.

In [ ]:
# Your code goes here. ################################################

Write your personal opinion on the activation functions, think about computation times too. Does BatchMeanSubtraction help?


In [ ]:
# Your answer goes here. ################################################

Finally, use all your knowledge to build a super cool model on this dataset, do not forget to split dataset into train and validation. Use dropout to prevent overfitting, play with learning rate decay. You can use data augmentation such as rotations, translations to boost your score. Use your knowledge and imagination to train a model.


In [ ]:
# Your code goes here. ################################################

Print here your accuracy. It should be around 90%.


In [ ]:
# Your answer goes here. ################################################

Autoencoder (optional)

This part is OPTIONAL, you may not do it. It will not be scored, but it is easy and interesting.

Now we are going to build a cool model, named autoencoder. The aim is simple: encode the data to a lower dimentional representation. Why? Well, if we can decode this representation back to original data with "small" reconstuction loss then we can store only compressed representation saving memory. But the most important thing is -- we can reuse trained autoencoder for classification.

Picture from this site.

Now implement an autoencoder:

Build it such that dimetionality inside autoencoder changes like that:

$$784 \text{ (data)} -> 512 -> 256 -> 128 -> 30 -> 128 -> 256 -> 512 -> 784$$

Use MSECriterion to score the reconstruction. Use BatchMeanNormalization between Linear and ReLU. You may not use nonlinearity in bottleneck layer.

You may train it for 9 epochs with batch size = 256, initial lr = 0.1 droping by a factor of 2 every 3 epochs. The reconstruction loss should be about 6.0 and visual quality decent already. Do not spend time on changing architecture, they are more or less the same.


In [ ]:
# Your code goes here. ################################################

Some time ago NNs were a lot poorer and people were struggling to learn deep models. To train a classification net people were training autoencoder first (to train autoencoder people were pretraining single layers with RBM), then substituting the decoder part with classification layer (yeah, they were struggling with training autoencoders a lot, and complex techniques were used at that dark times). We are going to this now, fast and easy.


In [ ]:
# Extract inner representation for train and validation, 
# you should get (n_samples, 30) matrices
# Your code goes here. ################################################

# Now build a logistic regression or small classification net
cnet = Sequential()
cnet.add(Linear(30, 2))
cnet.add(SoftMax())

# Learn the weights
# Your code goes here. ################################################

# Now chop off decoder part
# (you may need to implement `remove` method for Sequential container) 
# Your code goes here. ################################################

# And add learned layers ontop.
autoenc.add(cnet[0])
autoenc.add(cnet[1])

# Now optimize whole model
# Your code goes here. ################################################
  • What do you think, does it make sense to build real-world classifiers this way ? Did it work better for you than a straightforward one? Looks like it was not the same ~8 years ago, what has changed beside computational power?

Run PCA with 30 components on the train set, plot original image, autoencoder and PCA reconstructions side by side for 10 samples from validation set. Probably you need to use the following snippet to make aoutpencoder examples look comparible.


In [ ]:
# np.clip(prediction,0,1)
#
# Your code goes here. ################################################