Here is a Keras model of GoogLeNet (a.k.a Inception V1). I created it by converting the GoogLeNet model from Caffe. For more details on the conversion, see here.
GoogLeNet paper:
Going deeper with convolutions.
Szegedy, Christian, et al.
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015.
###Contents:
Custom layers for GoogLeNet: googlenet_custom_layers.py
Model and usage demo: googlenet.py
Weights: googlenet_weights.h5
| from scipy.misc import imread, imresize | |
| from keras.layers import Input, Dense, Convolution2D, MaxPooling2D, AveragePooling2D, ZeroPadding2D, Dropout, Flatten, merge, Reshape, Activation | |
| from keras.models import Model | |
| from keras.regularizers import l2 | |
| from keras.optimizers import SGD | |
| from googlenet_custom_layers import PoolHelper,LRN | |
| def create_googlenet(weights_path=None): | |
| # creates GoogLeNet a.k.a. Inception v1 (Szegedy, 2015) | |
| input = Input(shape=(3, 224, 224)) | |
| conv1_7x7_s2 = Convolution2D(64,7,7,subsample=(2,2),border_mode='same',activation='relu',name='conv1/7x7_s2',W_regularizer=l2(0.0002))(input) | |
| conv1_zero_pad = ZeroPadding2D(padding=(1, 1))(conv1_7x7_s2) | |
| pool1_helper = PoolHelper()(conv1_zero_pad) | |
| pool1_3x3_s2 = MaxPooling2D(pool_size=(3,3),strides=(2,2),border_mode='valid',name='pool1/3x3_s2')(pool1_helper) | |
| pool1_norm1 = LRN(name='pool1/norm1')(pool1_3x3_s2) | |
| conv2_3x3_reduce = Convolution2D(64,1,1,border_mode='same',activation='relu',name='conv2/3x3_reduce',W_regularizer=l2(0.0002))(pool1_norm1) | |
| conv2_3x3 = Convolution2D(192,3,3,border_mode='same',activation='relu',name='conv2/3x3',W_regularizer=l2(0.0002))(conv2_3x3_reduce) | |
| conv2_norm2 = LRN(name='conv2/norm2')(conv2_3x3) | |
| conv2_zero_pad = ZeroPadding2D(padding=(1, 1))(conv2_norm2) | |
| pool2_helper = PoolHelper()(conv2_zero_pad) | |
| pool2_3x3_s2 = MaxPooling2D(pool_size=(3,3),strides=(2,2),border_mode='valid',name='pool2/3x3_s2')(pool2_helper) | |
| inception_3a_1x1 = Convolution2D(64,1,1,border_mode='same',activation='relu',name='inception_3a/1x1',W_regularizer=l2(0.0002))(pool2_3x3_s2) | |
| inception_3a_3x3_reduce = Convolution2D(96,1,1,border_mode='same',activation='relu',name='inception_3a/3x3_reduce',W_regularizer=l2(0.0002))(pool2_3x3_s2) | |
| inception_3a_3x3 = Convolution2D(128,3,3,border_mode='same',activation='relu',name='inception_3a/3x3',W_regularizer=l2(0.0002))(inception_3a_3x3_reduce) | |
| inception_3a_5x5_reduce = Convolution2D(16,1,1,border_mode='same',activation='relu',name='inception_3a/5x5_reduce',W_regularizer=l2(0.0002))(pool2_3x3_s2) | |
| inception_3a_5x5 = Convolution2D(32,5,5,border_mode='same',activation='relu',name='inception_3a/5x5',W_regularizer=l2(0.0002))(inception_3a_5x5_reduce) | |
| inception_3a_pool = MaxPooling2D(pool_size=(3,3),strides=(1,1),border_mode='same',name='inception_3a/pool')(pool2_3x3_s2) | |
| inception_3a_pool_proj = Convolution2D(32,1,1,border_mode='same',activation='relu',name='inception_3a/pool_proj',W_regularizer=l2(0.0002))(inception_3a_pool) | |
| inception_3a_output = merge([inception_3a_1x1,inception_3a_3x3,inception_3a_5x5,inception_3a_pool_proj],mode='concat',concat_axis=1,name='inception_3a/output') | |
| inception_3b_1x1 = Convolution2D(128,1,1,border_mode='same',activation='relu',name='inception_3b/1x1',W_regularizer=l2(0.0002))(inception_3a_output) | |
| inception_3b_3x3_reduce = Convolution2D(128,1,1,border_mode='same',activation='relu',name='inception_3b/3x3_reduce',W_regularizer=l2(0.0002))(inception_3a_output) | |
| inception_3b_3x3 = Convolution2D(192,3,3,border_mode='same',activation='relu',name='inception_3b/3x3',W_regularizer=l2(0.0002))(inception_3b_3x3_reduce) | |
| inception_3b_5x5_reduce = Convolution2D(32,1,1,border_mode='same',activation='relu',name='inception_3b/5x5_reduce',W_regularizer=l2(0.0002))(inception_3a_output) | |
| inception_3b_5x5 = Convolution2D(96,5,5,border_mode='same',activation='relu',name='inception_3b/5x5',W_regularizer=l2(0.0002))(inception_3b_5x5_reduce) | |
| inception_3b_pool = MaxPooling2D(pool_size=(3,3),strides=(1,1),border_mode='same',name='inception_3b/pool')(inception_3a_output) | |
| inception_3b_pool_proj = Convolution2D(64,1,1,border_mode='same',activation='relu',name='inception_3b/pool_proj',W_regularizer=l2(0.0002))(inception_3b_pool) | |
| inception_3b_output = merge([inception_3b_1x1,inception_3b_3x3,inception_3b_5x5,inception_3b_pool_proj],mode='concat',concat_axis=1,name='inception_3b/output') | |
| inception_3b_output_zero_pad = ZeroPadding2D(padding=(1, 1))(inception_3b_output) | |
| pool3_helper = PoolHelper()(inception_3b_output_zero_pad) | |
| pool3_3x3_s2 = MaxPooling2D(pool_size=(3,3),strides=(2,2),border_mode='valid',name='pool3/3x3_s2')(pool3_helper) | |
| inception_4a_1x1 = Convolution2D(192,1,1,border_mode='same',activation='relu',name='inception_4a/1x1',W_regularizer=l2(0.0002))(pool3_3x3_s2) | |
| inception_4a_3x3_reduce = Convolution2D(96,1,1,border_mode='same',activation='relu',name='inception_4a/3x3_reduce',W_regularizer=l2(0.0002))(pool3_3x3_s2) | |
| inception_4a_3x3 = Convolution2D(208,3,3,border_mode='same',activation='relu',name='inception_4a/3x3',W_regularizer=l2(0.0002))(inception_4a_3x3_reduce) | |
| inception_4a_5x5_reduce = Convolution2D(16,1,1,border_mode='same',activation='relu',name='inception_4a/5x5_reduce',W_regularizer=l2(0.0002))(pool3_3x3_s2) | |
| inception_4a_5x5 = Convolution2D(48,5,5,border_mode='same',activation='relu',name='inception_4a/5x5',W_regularizer=l2(0.0002))(inception_4a_5x5_reduce) | |
| inception_4a_pool = MaxPooling2D(pool_size=(3,3),strides=(1,1),border_mode='same',name='inception_4a/pool')(pool3_3x3_s2) | |
| inception_4a_pool_proj = Convolution2D(64,1,1,border_mode='same',activation='relu',name='inception_4a/pool_proj',W_regularizer=l2(0.0002))(inception_4a_pool) | |
| inception_4a_output = merge([inception_4a_1x1,inception_4a_3x3,inception_4a_5x5,inception_4a_pool_proj],mode='concat',concat_axis=1,name='inception_4a/output') | |
| loss1_ave_pool = AveragePooling2D(pool_size=(5,5),strides=(3,3),name='loss1/ave_pool')(inception_4a_output) | |
| loss1_conv = Convolution2D(128,1,1,border_mode='same',activation='relu',name='loss1/conv',W_regularizer=l2(0.0002))(loss1_ave_pool) | |
| loss1_flat = Flatten()(loss1_conv) | |
| loss1_fc = Dense(1024,activation='relu',name='loss1/fc',W_regularizer=l2(0.0002))(loss1_flat) | |
| loss1_drop_fc = Dropout(0.7)(loss1_fc) | |
| loss1_classifier = Dense(1000,name='loss1/classifier',W_regularizer=l2(0.0002))(loss1_drop_fc) | |
| loss1_classifier_act = Activation('softmax')(loss1_classifier) | |
| inception_4b_1x1 = Convolution2D(160,1,1,border_mode='same',activation='relu',name='inception_4b/1x1',W_regularizer=l2(0.0002))(inception_4a_output) | |
| inception_4b_3x3_reduce = Convolution2D(112,1,1,border_mode='same',activation='relu',name='inception_4b/3x3_reduce',W_regularizer=l2(0.0002))(inception_4a_output) | |
| inception_4b_3x3 = Convolution2D(224,3,3,border_mode='same',activation='relu',name='inception_4b/3x3',W_regularizer=l2(0.0002))(inception_4b_3x3_reduce) | |
| inception_4b_5x5_reduce = Convolution2D(24,1,1,border_mode='same',activation='relu',name='inception_4b/5x5_reduce',W_regularizer=l2(0.0002))(inception_4a_output) | |
| inception_4b_5x5 = Convolution2D(64,5,5,border_mode='same',activation='relu',name='inception_4b/5x5',W_regularizer=l2(0.0002))(inception_4b_5x5_reduce) | |
| inception_4b_pool = MaxPooling2D(pool_size=(3,3),strides=(1,1),border_mode='same',name='inception_4b/pool')(inception_4a_output) | |
| inception_4b_pool_proj = Convolution2D(64,1,1,border_mode='same',activation='relu',name='inception_4b/pool_proj',W_regularizer=l2(0.0002))(inception_4b_pool) | |
| inception_4b_output = merge([inception_4b_1x1,inception_4b_3x3,inception_4b_5x5,inception_4b_pool_proj],mode='concat',concat_axis=1,name='inception_4b_output') | |
| inception_4c_1x1 = Convolution2D(128,1,1,border_mode='same',activation='relu',name='inception_4c/1x1',W_regularizer=l2(0.0002))(inception_4b_output) | |
| inception_4c_3x3_reduce = Convolution2D(128,1,1,border_mode='same',activation='relu',name='inception_4c/3x3_reduce',W_regularizer=l2(0.0002))(inception_4b_output) | |
| inception_4c_3x3 = Convolution2D(256,3,3,border_mode='same',activation='relu',name='inception_4c/3x3',W_regularizer=l2(0.0002))(inception_4c_3x3_reduce) | |
| inception_4c_5x5_reduce = Convolution2D(24,1,1,border_mode='same',activation='relu',name='inception_4c/5x5_reduce',W_regularizer=l2(0.0002))(inception_4b_output) | |
| inception_4c_5x5 = Convolution2D(64,5,5,border_mode='same',activation='relu',name='inception_4c/5x5',W_regularizer=l2(0.0002))(inception_4c_5x5_reduce) | |
| inception_4c_pool = MaxPooling2D(pool_size=(3,3),strides=(1,1),border_mode='same',name='inception_4c/pool')(inception_4b_output) | |
| inception_4c_pool_proj = Convolution2D(64,1,1,border_mode='same',activation='relu',name='inception_4c/pool_proj',W_regularizer=l2(0.0002))(inception_4c_pool) | |
| inception_4c_output = merge([inception_4c_1x1,inception_4c_3x3,inception_4c_5x5,inception_4c_pool_proj],mode='concat',concat_axis=1,name='inception_4c/output') | |
| inception_4d_1x1 = Convolution2D(112,1,1,border_mode='same',activation='relu',name='inception_4d/1x1',W_regularizer=l2(0.0002))(inception_4c_output) | |
| inception_4d_3x3_reduce = Convolution2D(144,1,1,border_mode='same',activation='relu',name='inception_4d/3x3_reduce',W_regularizer=l2(0.0002))(inception_4c_output) | |
| inception_4d_3x3 = Convolution2D(288,3,3,border_mode='same',activation='relu',name='inception_4d/3x3',W_regularizer=l2(0.0002))(inception_4d_3x3_reduce) | |
| inception_4d_5x5_reduce = Convolution2D(32,1,1,border_mode='same',activation='relu',name='inception_4d/5x5_reduce',W_regularizer=l2(0.0002))(inception_4c_output) | |
| inception_4d_5x5 = Convolution2D(64,5,5,border_mode='same',activation='relu',name='inception_4d/5x5',W_regularizer=l2(0.0002))(inception_4d_5x5_reduce) | |
| inception_4d_pool = MaxPooling2D(pool_size=(3,3),strides=(1,1),border_mode='same',name='inception_4d/pool')(inception_4c_output) | |
| inception_4d_pool_proj = Convolution2D(64,1,1,border_mode='same',activation='relu',name='inception_4d/pool_proj',W_regularizer=l2(0.0002))(inception_4d_pool) | |
| inception_4d_output = merge([inception_4d_1x1,inception_4d_3x3,inception_4d_5x5,inception_4d_pool_proj],mode='concat',concat_axis=1,name='inception_4d/output') | |
| loss2_ave_pool = AveragePooling2D(pool_size=(5,5),strides=(3,3),name='loss2/ave_pool')(inception_4d_output) | |
| loss2_conv = Convolution2D(128,1,1,border_mode='same',activation='relu',name='loss2/conv',W_regularizer=l2(0.0002))(loss2_ave_pool) | |
| loss2_flat = Flatten()(loss2_conv) | |
| loss2_fc = Dense(1024,activation='relu',name='loss2/fc',W_regularizer=l2(0.0002))(loss2_flat) | |
| loss2_drop_fc = Dropout(0.7)(loss2_fc) | |
| loss2_classifier = Dense(1000,name='loss2/classifier',W_regularizer=l2(0.0002))(loss2_drop_fc) | |
| loss2_classifier_act = Activation('softmax')(loss2_classifier) | |
| inception_4e_1x1 = Convolution2D(256,1,1,border_mode='same',activation='relu',name='inception_4e/1x1',W_regularizer=l2(0.0002))(inception_4d_output) | |
| inception_4e_3x3_reduce = Convolution2D(160,1,1,border_mode='same',activation='relu',name='inception_4e/3x3_reduce',W_regularizer=l2(0.0002))(inception_4d_output) | |
| inception_4e_3x3 = Convolution2D(320,3,3,border_mode='same',activation='relu',name='inception_4e/3x3',W_regularizer=l2(0.0002))(inception_4e_3x3_reduce) | |
| inception_4e_5x5_reduce = Convolution2D(32,1,1,border_mode='same',activation='relu',name='inception_4e/5x5_reduce',W_regularizer=l2(0.0002))(inception_4d_output) | |
| inception_4e_5x5 = Convolution2D(128,5,5,border_mode='same',activation='relu',name='inception_4e/5x5',W_regularizer=l2(0.0002))(inception_4e_5x5_reduce) | |
| inception_4e_pool = MaxPooling2D(pool_size=(3,3),strides=(1,1),border_mode='same',name='inception_4e/pool')(inception_4d_output) | |
| inception_4e_pool_proj = Convolution2D(128,1,1,border_mode='same',activation='relu',name='inception_4e/pool_proj',W_regularizer=l2(0.0002))(inception_4e_pool) | |
| inception_4e_output = merge([inception_4e_1x1,inception_4e_3x3,inception_4e_5x5,inception_4e_pool_proj],mode='concat',concat_axis=1,name='inception_4e/output') | |
| inception_4e_output_zero_pad = ZeroPadding2D(padding=(1, 1))(inception_4e_output) | |
| pool4_helper = PoolHelper()(inception_4e_output_zero_pad) | |
| pool4_3x3_s2 = MaxPooling2D(pool_size=(3,3),strides=(2,2),border_mode='valid',name='pool4/3x3_s2')(pool4_helper) | |
| inception_5a_1x1 = Convolution2D(256,1,1,border_mode='same',activation='relu',name='inception_5a/1x1',W_regularizer=l2(0.0002))(pool4_3x3_s2) | |
| inception_5a_3x3_reduce = Convolution2D(160,1,1,border_mode='same',activation='relu',name='inception_5a/3x3_reduce',W_regularizer=l2(0.0002))(pool4_3x3_s2) | |
| inception_5a_3x3 = Convolution2D(320,3,3,border_mode='same',activation='relu',name='inception_5a/3x3',W_regularizer=l2(0.0002))(inception_5a_3x3_reduce) | |
| inception_5a_5x5_reduce = Convolution2D(32,1,1,border_mode='same',activation='relu',name='inception_5a/5x5_reduce',W_regularizer=l2(0.0002))(pool4_3x3_s2) | |
| inception_5a_5x5 = Convolution2D(128,5,5,border_mode='same',activation='relu',name='inception_5a/5x5',W_regularizer=l2(0.0002))(inception_5a_5x5_reduce) | |
| inception_5a_pool = MaxPooling2D(pool_size=(3,3),strides=(1,1),border_mode='same',name='inception_5a/pool')(pool4_3x3_s2) | |
| inception_5a_pool_proj = Convolution2D(128,1,1,border_mode='same',activation='relu',name='inception_5a/pool_proj',W_regularizer=l2(0.0002))(inception_5a_pool) | |
| inception_5a_output = merge([inception_5a_1x1,inception_5a_3x3,inception_5a_5x5,inception_5a_pool_proj],mode='concat',concat_axis=1,name='inception_5a/output') | |
| inception_5b_1x1 = Convolution2D(384,1,1,border_mode='same',activation='relu',name='inception_5b/1x1',W_regularizer=l2(0.0002))(inception_5a_output) | |
| inception_5b_3x3_reduce = Convolution2D(192,1,1,border_mode='same',activation='relu',name='inception_5b/3x3_reduce',W_regularizer=l2(0.0002))(inception_5a_output) | |
| inception_5b_3x3 = Convolution2D(384,3,3,border_mode='same',activation='relu',name='inception_5b/3x3',W_regularizer=l2(0.0002))(inception_5b_3x3_reduce) | |
| inception_5b_5x5_reduce = Convolution2D(48,1,1,border_mode='same',activation='relu',name='inception_5b/5x5_reduce',W_regularizer=l2(0.0002))(inception_5a_output) | |
| inception_5b_5x5 = Convolution2D(128,5,5,border_mode='same',activation='relu',name='inception_5b/5x5',W_regularizer=l2(0.0002))(inception_5b_5x5_reduce) | |
| inception_5b_pool = MaxPooling2D(pool_size=(3,3),strides=(1,1),border_mode='same',name='inception_5b/pool')(inception_5a_output) | |
| inception_5b_pool_proj = Convolution2D(128,1,1,border_mode='same',activation='relu',name='inception_5b/pool_proj',W_regularizer=l2(0.0002))(inception_5b_pool) | |
| inception_5b_output = merge([inception_5b_1x1,inception_5b_3x3,inception_5b_5x5,inception_5b_pool_proj],mode='concat',concat_axis=1,name='inception_5b/output') | |
| pool5_7x7_s1 = AveragePooling2D(pool_size=(7,7),strides=(1,1),name='pool5/7x7_s2')(inception_5b_output) | |
| loss3_flat = Flatten()(pool5_7x7_s1) | |
| pool5_drop_7x7_s1 = Dropout(0.4)(loss3_flat) | |
| loss3_classifier = Dense(1000,name='loss3/classifier',W_regularizer=l2(0.0002))(pool5_drop_7x7_s1) | |
| loss3_classifier_act = Activation('softmax',name='prob')(loss3_classifier) | |
| googlenet = Model(input=input, output=[loss1_classifier_act,loss2_classifier_act,loss3_classifier_act]) | |
| if weights_path: | |
| googlenet.load_weights(weights_path) | |
| return googlenet | |
| if __name__ == "__main__": | |
| img = imresize(imread('cat.jpg', mode='RGB'), (224, 224)).astype(np.float32) | |
| img[:, :, 0] -= 123.68 | |
| img[:, :, 1] -= 116.779 | |
| img[:, :, 2] -= 103.939 | |
| img[:,:,[0,1,2]] = img[:,:,[2,1,0]] | |
| img = img.transpose((2, 0, 1)) | |
| img = np.expand_dims(img, axis=0) | |
| # Test pretrained model | |
| model = create_googlenet('googlenet_weights.h5') | |
| sgd = SGD(lr=0.1, decay=1e-6, momentum=0.9, nesterov=True) | |
| model.compile(optimizer=sgd, loss='categorical_crossentropy') | |
| out = model.predict(img) # note: the model has three outputs | |
| print np.argmax(out[2]) |
| from keras.layers.core import Layer | |
| import theano.tensor as T | |
| class LRN(Layer): | |
| def __init__(self, alpha=0.0001,k=1,beta=0.75,n=5, **kwargs): | |
| self.alpha = alpha | |
| self.k = k | |
| self.beta = beta | |
| self.n = n | |
| super(LRN, self).__init__(**kwargs) | |
| def call(self, x, mask=None): | |
| b, ch, r, c = x.shape | |
| half_n = self.n // 2 # half the local region | |
| input_sqr = T.sqr(x) # square the input | |
| extra_channels = T.alloc(0., b, ch + 2*half_n, r, c) # make an empty tensor with zero pads along channel dimension | |
| input_sqr = T.set_subtensor(extra_channels[:, half_n:half_n+ch, :, :],input_sqr) # set the center to be the squared input | |
| scale = self.k # offset for the scale | |
| norm_alpha = self.alpha / self.n # normalized alpha | |
| for i in range(self.n): | |
| scale += norm_alpha * input_sqr[:, i:i+ch, :, :] | |
| scale = scale ** self.beta | |
| x = x / scale | |
| return x | |
| def get_config(self): | |
| config = {"alpha": self.alpha, | |
| "k": self.k, | |
| "beta": self.beta, | |
| "n": self.n} | |
| base_config = super(LRN, self).get_config() | |
| return dict(list(base_config.items()) + list(config.items())) | |
| class PoolHelper(Layer): | |
| def __init__(self, **kwargs): | |
| super(PoolHelper, self).__init__(**kwargs) | |
| def call(self, x, mask=None): | |
| return x[:,:,1:,1:] | |
| def get_config(self): | |
| config = {} | |
| base_config = super(PoolHelper, self).get_config() | |
| return dict(list(base_config.items()) + list(config.items())) | |