def check_fast_conv(self):
x = np.random.randn(100, 3, 31, 31)
w = np.random.randn(25, 3, 3, 3)
b = np.random.randn(25,)
dout = np.random.randn(100, 25, 16, 16)
conv_param = {'stride': 2, 'pad': 1}
t0 = time()
out_naive, cache_naive = conv_forward_naive(x, w, b, conv_param)
t1 = time()
out_fast, cache_fast = conv_forward_fast(x, w, b, conv_param)
t2 = time()
print 'Testing conv_forward_fast:'
print 'Naive: %fs' % (t1 - t0)
print 'Fast: %fs' % (t2 - t1)
print 'Speedup: %fx' % ((t1 - t0) / (t2 - t1))
print 'Difference: ', self.rel_error(out_naive, out_fast)
t0 = time()
dx_naive, dw_naive, db_naive = conv_backward_naive(dout, cache_naive)
t1 = time()
dx_fast, dw_fast, db_fast = conv_backward_fast(dout, cache_fast)
t2 = time()
print '\nTesting conv_backward_fast:'
print 'Naive: %fs' % (t1 - t0)
print 'Fast: %fs' % (t2 - t1)
print 'Speedup: %fx' % ((t1 - t0) / (t2 - t1))
print 'dx difference: ', self.rel_error(dx_naive, dx_fast)
print 'dw difference: ', self.rel_error(dw_naive, dw_fast)
print 'db difference: ', self.rel_error(db_naive, db_fast)
return
def conv_relu_forward(x, w, b, conv_param):
"""
A convenience layer that performs a convolution followed by a ReLU.
Inputs:
- x: Input to the convolutional layer
- w, b, conv_param: Weights and parameters for the convolutional layer
Returns a tuple of:
- out: Output from the ReLU
- cache: Object to give to the backward pass
"""
a, conv_cache = conv_forward_fast(x, w, b, conv_param)
out, relu_cache = relu_forward(a)
cache = (conv_cache, relu_cache)
return out, cache
def conv_relu_forward(x, w, b, conv_param):
"""
A convenience layer that performs a convolution followed by a ReLU.
Inputs:
- x: Input to the convolutional layer
- w, b, conv_param: Weights and parameters for the convolutional layer
Returns a tuple of:
- out: Output from the ReLU
- cache: Object to give to the backward pass
"""
a, conv_cache = conv_forward_fast(x, w, b, conv_param)
out, relu_cache = relu_forward(a)
cache = (conv_cache, relu_cache)
return out, cache
def blur_image(X):
"""
A very gentle image blurring operation, to be used as a regularizer for image
generation.
Inputs:
- X: Image data of shape (N, 3, H, W)
Returns:
- X_blur: Blurred version of X, of shape (N, 3, H, W)
"""
w_blur = np.zeros((3, 3, 3, 3))
b_blur = np.zeros(3)
blur_param = {'stride': 1, 'pad': 1}
for i in xrange(3):
w_blur[i, i] = np.asarray([[1, 2, 1], [2, 188, 2], [1, 2, 1]], dtype=np.float32)
w_blur /= 200.0
return conv_forward_fast(X, w_blur, b_blur, blur_param)[0]
def conv_relu_pool_forward(x, w, b, conv_param, pool_param):
"""
Convenience layer that performs a convolution, a ReLU, and a pool.
Inputs:
- x: Input to the convolutional layer
- w, b, conv_param: Weights and parameters for the convolutional layer
- pool_param: Parameters for the pooling layer
Returns a tuple of:
- out: Output from the pooling layer
- cache: Object to give to the backward pass
"""
a, conv_cache = conv_forward_fast(x, w, b, conv_param)
s, relu_cache = relu_forward(a)
out, pool_cache = max_pool_forward_fast(s, pool_param)
cache = (conv_cache, relu_cache, pool_cache)
return out, cache
def blur_image(X):
"""
A very gentle image blurring operation, to be used as a regularizer for image
generation.
Inputs:
- X: Image data of shape (N, 3, H, W)
Returns:
- X_blur: Blurred version of X, of shape (N, 3, H, W)
"""
w_blur = np.zeros((3, 3, 3, 3))
b_blur = np.zeros(3)
blur_param = {'stride': 1, 'pad': 1}
for i in range(3):
w_blur[i, i] = np.asarray([[1, 2, 1], [2, 188, 2], [1, 2, 1]], dtype=np.float32)
w_blur /= 200.0
return conv_forward_fast(X, w_blur, b_blur, blur_param)[0]
def conv_relu_pool_forward(x, w, b, conv_param, pool_param):
"""
Convenience layer that performs a convolution, a ReLU, and a pool.
Inputs:
- x: Input to the convolutional layer
- w, b, conv_param: Weights and parameters for the convolutional layer
- pool_param: Parameters for the pooling layer
Returns a tuple of:
- out: Output from the pooling layer
- cache: Object to give to the backward pass
"""
a, conv_cache = conv_forward_fast(x, w, b, conv_param)
s, relu_cache = relu_forward(a)
out, pool_cache = max_pool_forward_fast(s, pool_param)
cache = (conv_cache, relu_cache, pool_cache)
return out, cache
def blur_image(X):
"""
模糊操作，用于图像生成
Inputs:
- X: 图像，形状是 (N, 3, H, W)
Returns:
- X_blur: 模糊的 X, 形状是 (N, 3, H, W)
"""
from cs231n.fast_layers import conv_forward_fast
w_blur = np.zeros((3, 3, 3, 3))
b_blur = np.zeros(3)
blur_param = {'stride': 1, 'pad': 1}
for i in range(3):
w_blur[i, i] = np.asarray([[1, 2, 1], [2, 188, 2], [1, 2, 1]],
dtype=np.float32)
w_blur /= 200.0
return conv_forward_fast(X, w_blur, b_blur, blur_param)[0]
# by running the following:
# Rel errors should be around e-9 or less
from cs231n.fast_layers import conv_forward_fast, conv_backward_fast
from time import time
np.random.seed(231)
x = np.random.randn(100, 3, 31, 31)
w = np.random.randn(25, 3, 3, 3)
b = np.random.randn(25, )
dout = np.random.randn(100, 25, 16, 16)
conv_param = {'stride': 2, 'pad': 1}
t0 = time()
out_naive, cache_naive = conv_forward_naive(x, w, b, conv_param)
t1 = time()
out_fast, cache_fast = conv_forward_fast(x, w, b, conv_param)
t2 = time()
print('Testing conv_forward_fast:')
print('Naive: %fs' % (t1 - t0))
print('Fast: %fs' % (t2 - t1))
print('Speedup: %fx' % ((t1 - t0) / (t2 - t1)))
print('Difference: ', rel_error(out_naive, out_fast))
t0 = time()
dx_naive, dw_naive, db_naive = conv_backward_naive(dout, cache_naive)
t1 = time()
dx_fast, dw_fast, db_fast = conv_backward_fast(dout, cache_fast)
t2 = time()
print('\nTesting conv_backward_fast:')
def _forward(self, x):
global params
return conv_forward_fast(x, params[self.n('w')], params[self.n('b')],
self.conv_param)
def loss(self, X, y=None):
"""
Evaluate loss and gradient for the three-layer convolutional network.
Input / output: Same API as TwoLayerNet in fc_net.py.
"""
W1, b1 = self.params['W1'], self.params['b1']
W2, b2 = self.params['W2'], self.params['b2']
W3, b3 = self.params['W3'], self.params['b3']
# pass conv_param to the forward pass for the convolutional layer
filter_size = W1.shape[2]
conv_param = {'stride': 1, 'pad': (filter_size - 1) / 2}
# pass pool_param to the forward pass for the max-pooling layer
pool_param = {'pool_height': 2, 'pool_width': 2, 'stride': 2}
#conv
#conv - relu - 2x2 max pool - affine - relu - affine - softmax
convout = None
print "ipiut", X.shape
convout, self.cache['Input'] = conv_forward_fast(X, W1, b1, conv_param)
print "conv out", convout.shape
relu1, self.cache['relu1'] = relu_forward(convout)
print "relu out", relu1.shape
maxpoolout, self.cache['maxpool'] = max_pool_forward_fast(
relu1, pool_param)
print "maxpool out", maxpoolout.shape
affineout1, self.cache['affine1'] = affine_forward(maxpoolout, W2, b2)
print "affine out", affineout1.shape
reluout2, self.cache['relu2'] = relu_forward(affineout1)
print "relu2", reluout2.shape
affineout2, self.cache['affine2'] = affine_forward(reluout2, W3, b3)
print "affine out2", affineout2.shape
############################################################################
# TODO: Implement the forward pass for the three-layer convolutional net, #
# computing the class scores for X and storing them in the scores #
# variable. #
############################################################################
pass
############################################################################s
# END OF YOUR CODE #
############################################################################
if y is None:
return scores
loss, grads = 0, {}
loss, dx = softmax_loss(affineout2, y)
grads['W3'] = np.dot(reluout2.T, dx)
grads['b3'] = np.sum(dx, axis=0)
dx[affineout2 <= 0] = 0
#relu2back=relu_backward(drelu2,self.cache['relu2'])
print "W2 shape:", self.params['W2'].shape
affineback1, dw, db = affine_backward(
dx, [reluout2, self.params['W3'], self.params['b3']])
grads['W2'] = np.dot(maxpoolout.T, affineback1)
grads['b2'] = np.sum(affineback1, axis=0)
grads['W1'] = self.params['W1']
grads['b1'] = self.params['b1']
############################################################################
# TODO: Implement the backward pass for the three-layer convolutional net, #
# storing the loss and gradients in the loss and grads variables. 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! #
############################################################################
pass
############################################################################
# END OF YOUR CODE #
############################################################################
return loss, grads
""" returns relative error """
return np.max(np.abs(x - y) / (np.maximum(1e-8, np.abs(x) + np.abs(y))))
from cs231n.fast_layers import conv_forward_fast, conv_backward_fast
from time import time
x = np.random.randn(100, 3, 31, 31)
w = np.random.randn(25, 3, 3, 3)
b = np.random.randn(25,)
dout = np.random.randn(100, 25, 16, 16)
conv_param = {'stride': 2, 'pad': 1}
t0 = time()
out_naive, cache_naive = conv_forward_naive(x, w, b, conv_param)
t1 = time()
out_fast, cache_fast = conv_forward_fast(x, w, b, conv_param)
t2 = time()
print 'Testing conv_forward_fast:'
print 'Naive: %fs' % (t1 - t0)
print 'Fast: %fs' % (t2 - t1)
print 'Speedup: %fx' % ((t1 - t0) / (t2 - t1))
print 'Difference: ', rel_error(out_naive, out_fast)
t0 = time()
dx_naive, dw_naive, db_naive = conv_backward_naive(dout, cache_naive)
t1 = time()
dx_fast, dw_fast, db_fast = conv_backward_fast(dout, cache_fast)
t2 = time()
print '\nTesting conv_backward_fast:'