assignment2/cs231n/classifiers/fc_net.py

from builtins import range
from builtins import object
import numpy as np

from cs231n.layers import *
from cs231n.layer_utils import *


class TwoLayerNet(object):
    """
    A two-layer fully-connected neural network with ReLU nonlinearity and
    softmax loss that uses a modular layer design. We assume an input dimension
    of D, a hidden dimension of H, and perform classification over C classes.

    The architecure should be affine - relu - affine - softmax.

    Note that this class does not implement gradient descent; instead, it
    will interact with a separate Solver object that is responsible for running
    optimization.

    The learnable parameters of the model are stored in the dictionary
    self.params that maps parameter names to numpy arrays.
    """

    def __init__(self, input_dim=3*32*32, hidden_dim=100, num_classes=10,
                 weight_scale=1e-3, reg=0.0):
        """
        Initialize a new network.

        Inputs:
        - input_dim: An integer giving the size of the input
        - hidden_dim: An integer giving the size of the hidden layer
        - num_classes: An integer giving the number of classes to classify
        - weight_scale: Scalar giving the standard deviation for random
          initialization of the weights.
        - reg: Scalar giving L2 regularization strength.
        """
        self.params = {}
        self.reg = reg

        ############################################################################
        # TODO: Initialize the weights and biases of the two-layer net. Weights    #
        # should be initialized from a Gaussian centered at 0.0 with               #
        # standard deviation equal to weight_scale, and biases should be           #
        # initialized to zero. All weights and biases should be stored in the      #
        # dictionary self.params, with first layer weights                         #
        # and biases using the keys 'W1' and 'b1' and second layer                 #
        # weights and biases using the keys 'W2' and 'b2'.                         #
        ############################################################################
        # *****START OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****

        self.params['W1'] = weight_scale * np.random.randn(input_dim, hidden_dim)
        self.params['b1'] = np.zeros(hidden_dim)
        self.params['W2'] = weight_scale * np.random.randn(hidden_dim, num_classes)
        self.params['b2'] = np.zeros(num_classes)


        # *****END OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****
        ############################################################################
        #                             END OF YOUR CODE                             #
        ############################################################################


    def loss(self, X, y=None):
        """
        Compute loss and gradient for a minibatch of data.

        Inputs:
        - X: Array of input data of shape (N, d_1, ..., d_k)
        - y: Array of labels, of shape (N,). y[i] gives the label for X[i].

        Returns:
        If y is None, then run a test-time forward pass of the model and return:
        - scores: Array of shape (N, C) giving classification scores, where
          scores[i, c] is the classification score for X[i] and class c.

        If y is not None, then run a training-time forward and backward pass and
        return a tuple of:
        - loss: Scalar value giving the loss
        - grads: Dictionary with the same keys as self.params, mapping parameter
          names to gradients of the loss with respect to those parameters.
        """
        scores = None
        ############################################################################
        # TODO: Implement the forward pass for the two-layer net, computing the    #
        # class scores for X and storing them in the scores variable.              #
        ############################################################################
        # *****START OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****

        W1, b1 = self.params['W1'], self.params['b1']
        W2, b2 = self.params['W2'], self.params['b2']

        # affine_relu_layer
        out1, cache1 = affine_relu_forward(X, W1, b1)
        # affine_layer
        out2, cache2 = affine_forward(out1, W2, b2)
        scores = out2

        # *****END OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****
        ############################################################################
        #                             END OF YOUR CODE                             #
        ############################################################################

        # If y is None then we are in test mode so just return scores
        if y is None:
            return scores

        loss, grads = 0, {}
        ############################################################################
        # TODO: Implement the backward pass for the two-layer net. Store the loss  #
        # in the loss variable and gradients in the grads dictionary. Compute data #
        # loss using softmax, and make sure that grads[k] holds the gradients for  #
        # self.params[k]. Don't forget to add L2 regularization!                   #
        #                                                                          #
        # NOTE: To ensure that your implementation matches ours and you pass the   #
        # automated tests, make sure that your L2 regularization includes a factor #
        # of 0.5 to simplify the expression for the gradient.                      #
        ############################################################################
        # *****START OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****

        # compute loss
        data_loss, dscores = softmax_loss(scores, y)
        reg_loss = 0.5 * self.reg * (np.sum(W1 * W1) + np.sum(W2 * W2))
        loss = data_loss + reg_loss

        # compute grads
        dhidden, grads['W2'], grads['b2'] = affine_backward(dscores, cache2)
        grads['W2'] += self.reg * W2

        dX, grads['W1'], grads['b1'] = affine_relu_backward(dhidden, cache1)
        grads['W1'] += self.reg * W1

        # *****END OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****
        ############################################################################
        #                             END OF YOUR CODE                             #
        ############################################################################

        return loss, grads


class FullyConnectedNet(object):
    """
    A fully-connected neural network with an arbitrary number of hidden layers,
    ReLU nonlinearities, and a softmax loss function. This will also implement
    dropout and batch/layer normalization as options. For a network with L layers,
    the architecture will be

    {affine - [batch/layer norm] - relu - [dropout]} x (L - 1) - affine - softmax

    where batch/layer normalization and dropout are optional, and the {...} block is
    repeated L - 1 times.

    Similar to the TwoLayerNet above, learnable parameters are stored in the
    self.params dictionary and will be learned using the Solver class.
    """

    def __init__(self, hidden_dims, input_dim=3*32*32, num_classes=10,
                 dropout=1, normalization=None, reg=0.0,
                 weight_scale=1e-2, dtype=np.float32, seed=None):
        """
        Initialize a new FullyConnectedNet.

        Inputs:
        - hidden_dims: A list of integers giving the size of each hidden layer.
        - input_dim: An integer giving the size of the input.
        - num_classes: An integer giving the number of classes to classify.
        - dropout: Scalar between 0 and 1 giving dropout strength. If dropout=1 then
          the network should not use dropout at all.
        - normalization: What type of normalization the network should use. Valid values
          are "batchnorm", "layernorm", or None for no normalization (the default).
        - reg: Scalar giving L2 regularization strength.
        - weight_scale: Scalar giving the standard deviation for random
          initialization of the weights.
        - dtype: A numpy datatype object; all computations will be performed using
          this datatype. float32 is faster but less accurate, so you should use
          float64 for numeric gradient checking.
        - seed: If not None, then pass this random seed to the dropout layers. This
          will make the dropout layers deteriminstic so we can gradient check the
          model.
        """
        self.normalization = normalization
        self.use_dropout = dropout != 1
        self.reg = reg
        self.num_layers = 1 + len(hidden_dims)
        self.dtype = dtype
        self.params = {}

        ############################################################################
        # TODO: Initialize the parameters of the network, storing all values in    #
        # the self.params dictionary. Store weights and biases for the first layer #
        # in W1 and b1; for the second layer use W2 and b2, etc. Weights should be #
        # initialized from a normal distribution centered at 0 with standard       #
        # deviation equal to weight_scale. Biases should be initialized to zero.   #
        #                                                                          #
        # When using batch normalization, store scale and shift parameters for the #
        # first layer in gamma1 and beta1; for the second layer use gamma2 and     #
        # beta2, etc. Scale parameters should be initialized to ones and shift     #
        # parameters should be initialized to zeros.                               #
        ############################################################################
        # *****START OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****

        layer_dims = [input_dim] + hidden_dims + [num_classes]
        for l in range(len(layer_dims) - 1):
            str_W = 'W' + str('%d' % (l + 1))
            str_b = 'b' + str('%d' % (l + 1))
            self.params[str_W] = weight_scale * np.random.randn(layer_dims[l], layer_dims[l + 1])
            self.params[str_b] = np.zeros(layer_dims[l + 1])

            if self.normalization is not None:
                str_gamma = 'gamma' + str('%d' % (l + 1))
                str_beta = 'beta' + str('%d' % (l + 1))
                self.params[str_gamma] = np.ones(layer_dims[l + 1])
                self.params[str_beta] = np.zeros(layer_dims[l + 1])

        # Note: number of param:gamma and beta is less 1 than w and b
        # Delete the last gamma and beta, because their num is less one than W and b
        if self.normalization is not None:
            str_gamma = 'gamma' + str('%d' % (self.num_layers))
            str_beta = 'beta' + str('%d' % (self.num_layers))
            del self.params[str_gamma], self.params[str_beta]


        # *****END OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****
        ############################################################################
        #                             END OF YOUR CODE                             #
        ############################################################################

        # When using dropout we need to pass a dropout_param dictionary to each
        # dropout layer so that the layer knows the dropout probability and the mode
        # (train / test). You can pass the same dropout_param to each dropout layer.
        self.dropout_param = {}
        if self.use_dropout:
            self.dropout_param = {'mode': 'train', 'p': dropout}
            if seed is not None:
                self.dropout_param['seed'] = seed

        # With batch normalization we need to keep track of running means and
        # variances, so we need to pass a special bn_param object to each batch
        # normalization layer. You should pass self.bn_params[0] to the forward pass
        # of the first batch normalization layer, self.bn_params[1] to the forward
        # pass of the second batch normalization layer, etc.
        self.bn_params = []
        if self.normalization=='batchnorm':
            self.bn_params = [{'mode': 'train'} for i in range(self.num_layers - 1)]
        if self.normalization=='layernorm':
            self.bn_params = [{} for i in range(self.num_layers - 1)]

        # Cast all parameters to the correct datatype
        for k, v in self.params.items():
            self.params[k] = v.astype(dtype)


    def loss(self, X, y=None):
        """
        Compute loss and gradient for the fully-connected net.

        Input / output: Same as TwoLayerNet above.
        """
        X = X.astype(self.dtype)
        mode = 'test' if y is None else 'train'

        # Set train/test mode for batchnorm params and dropout param since they
        # behave differently during training and testing.
        if self.use_dropout:
            self.dropout_param['mode'] = mode
        if self.normalization=='batchnorm':
            for bn_param in self.bn_params:
                bn_param['mode'] = mode
        scores = None
        ############################################################################
        # TODO: Implement the forward pass for the fully-connected net, computing  #
        # the class scores for X and storing them in the scores variable.          #
        #                                                                          #
        # When using dropout, you'll need to pass self.dropout_param to each       #
        # dropout forward pass.                                                    #
        #                                                                          #
        # When using batch normalization, you'll need to pass self.bn_params[0] to #
        # the forward pass for the first batch normalization layer, pass           #
        # self.bn_params[1] to the forward pass for the second batch normalization #
        # layer, etc.                                                              #
        ############################################################################
        # *****START OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****

        # Now I just write code without batch/layer norm and dropout layer--2019/8/18
        # Today I add dropout layer after ReLU--2019/8/19
        # Today I add batch normalization layer before ReLU--2019/8/20
        XX = X.reshape(X.shape[0], -1).copy()
        N, D = XX.shape

        # the params between input layer and first hidden layer
        hidden_in_value = [] # It seems no need
        hidden_mid_value = []
        hidden_out_value = []

        cache_af_value = []
        cache_relu_value = []
        cache_bn_value = []
        cache_ln_value = []
        cache_dropout_value = []

        hidden_out = X
        # the forward pass between input layer/hidden layer and hidden layer
        num_hidden_layers = self.num_layers - 1
        for i in range(num_hidden_layers):
            str_W = 'W' + str('%d' %(i+1))
            str_b = 'b' + str('%d' %(i+1))
            str_gamma = 'gamma' + str('%d' %(i+1))
            str_beta = 'beta' + str('%d' %(i+1))

            hidden_in, cache_af = affine_forward(hidden_out, self.params[str_W], self.params[str_b])
            hidden_in_value.append(hidden_in) # recorde forward pass value for backprop,not used later
            cache_af_value.append(cache_af)
            if self.normalization is not None:
                if self.normalization == 'batchnorm':
                    hidden_norm, cache_bn = batchnorm_forward(hidden_in, self.params[str_gamma],
                    self.params[str_beta], self.bn_params[i]) # do not forget gamma & beta
                    cache_bn_value.append(cache_bn)
                    
                if self.normalization == 'layernorm':
                    hidden_norm, cache_ln = layernorm_forward(hidden_in, self.params[str_gamma], 
                    self.params[str_beta], self.bn_params[i])
                    cache_ln_value.append(cache_ln)
            else:
                hidden_norm = hidden_in

            hidden_mid, cache_relu = relu_forward(hidden_norm)
            hidden_mid_value.append(hidden_mid)
            cache_relu_value.append(cache_relu)

            if self.use_dropout:
                hidden_out, cache_dropout = dropout_forward(hidden_mid, self.dropout_param)
                cache_dropout_value.append(cache_dropout)
            else:
                hidden_out = hidden_mid
            hidden_out_value.append(hidden_out) # recorde forward pass value for backprop

        # the params between last hidden layer and output layer, no gamma and beta
        str_W = 'W' + str('%d' %(num_hidden_layers + 1))
        str_b = 'b' + str('%d' %(num_hidden_layers + 1))
        scores = hidden_out.dot(self.params[str_W]) + self.params[str_b]

        # *****END OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****
        ############################################################################
        #                             END OF YOUR CODE                             #
        ############################################################################

        # If test mode return early
        if mode == 'test':
            return scores

        loss, grads = 0.0, {}
        ############################################################################
        # TODO: Implement the backward pass for the fully-connected net. Store the #
        # loss in the loss variable and gradients in the grads dictionary. Compute #
        # data loss using softmax, and make sure that grads[k] holds the gradients #
        # for self.params[k]. Don't forget to add L2 regularization!               #
        #                                                                          #
        # When using batch/layer normalization, you don't need to regularize the   #
        # scale and shift parameters.                                              #
        #                                                                          #
        # NOTE: To ensure that your implementation matches ours and you pass the   #
        # automated tests, make sure that your L2 regularization includes a factor #
        # of 0.5 to simplify the expression for the gradient.                      #
        ############################################################################
        # *****START OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****

        # Now I just write code without batch/layer norm and dropout layer--2019/8/18
        ############################ compute loss ##################################
        data_loss, dscores = softmax_loss(scores, y)
        reg_loss = 0.0
        for i in range(self.num_layers):
            str_W = 'W' + str('%d' %(i+1))
            reg_loss += np.sum(self.params[str_W] ** 2)

        reg_loss = 0.5 * self.reg * reg_loss
        loss = data_loss + reg_loss

        ############################ compute grads #################################
        str_W = 'W' + str('%d' % self.num_layers)
        str_b = 'b' + str('%d' % self.num_layers)
        grads[str_W] = np.dot(hidden_out.T, dscores) + self.reg * self.params[str_W]
        grads[str_b] = np.sum(dscores, axis=0)

        grad_x_in = np.dot(dscores, self.params[str_W].T)

        for i in range(num_hidden_layers, 0, -1):

            if self.use_dropout:
                grad_x_mid = dropout_backward(grad_x_in, cache_dropout_value[i-1])
            else:
                grad_x_mid = grad_x_in

            # Pass ReLU
            grad_x_norm = relu_backward(grad_x_mid, cache_relu_value[i-1])

            str_gamma = 'gamma' + str('%d' % i)
            str_beta = 'beta' + str('%d' % i)
            if self.normalization is not None:
                if self.normalization == 'batchnorm':
                    grad_x_out, grads[str_gamma], grads[str_beta] = \
                        batchnorm_backward_alt(grad_x_norm, cache_bn_value[i-1]) # Do not forget dgamma, dbeta
                if self.normalization == 'layernorm':
                    grad_x_out, grads[str_gamma], grads[str_beta] = \
                        layernorm_backward(grad_x_norm, cache_ln_value[i-1])
            else:
                grad_x_out = grad_x_norm

            str_W = 'W' + str('%d' % i)
            str_b = 'b' + str('%d' % i)

            grad_x_in, grads[str_W], grads[str_b] = affine_backward(grad_x_out, cache_af_value[i-1])
            grads[str_W] += self.reg * self.params[str_W]


        # *****END OF YOUR CODE (DO NOT DELETE/MODIFY THIS LINE)*****
        ############################################################################
        #                             END OF YOUR CODE                             #
        ############################################################################

        return loss, grads