joelouismarino

img = imread('cat.jpg', mode='RGB')
height,width = img.shape[:2]
img = img.astype('float32')
# subtract means
img[:, :, 0] -= 123.68
img[:, :, 1] -= 116.779
img[:, :, 2] -= 103.939
img[:,:,[0,1,2]] = img[:, :, [2, 1, 0]] # swap channels
img = img.transpose((2, 0, 1)) # re-order dimensions
img = img[:,(height-224)//2:(height+224)//2, (width-224)//2:(width+224)//2] #crop

joelouismarino

if 'conv' in layer_name or 'proj' in layer_name or (('1x1' in layer_name or '3x3' in layer_name or '5x5' in layer_name) and 'inception' in layer_name):
    for i in range(weights.shape[0]): # go through each filter
    for j in range(weights.shape[1]): # go through each channel
        weights[i, j] = np.rot90(weights[i, j], 2) # rotate it (twice)

joelouismarino

if 'fc' in layer_name or 'classifier' in layer_name:
	weights = np.transpose(weights)

joelouismarino

for layer_name in net.params.keys():
    weights = np.copy(net.params[layer_name][0].data)
    biases = np.copy(net.params[layer_name][1].data)
    model_layer = googlenet.get_layer(name=layer_name)
    model_layer.set_weights([weights, biases])

joelouismarino

import caffe

MODEL_DEF = 'path/to/train_val.prototxt'
MODEL_WEIGHTS = 'path/to/bvlc_googlenet.caffemodel'
net = caffe.Net(MODEL_DEF, MODEL_WEIGHTS, caffe.TEST)

joelouismarino

googlenet = Model(input=input, output=[loss1_classifier_act, loss2_classifier_act, loss3_classifier_act])

joelouismarino

loss3_classifier = Dense(1000, name='loss3/classifier')(pool5_drop_7x7_s1)
loss3_classifier_act = Activation('softmax', name='prob')(loss3_classifier)

joelouismarino

layer {
    name: "loss3/classifier"
    type: "InnerProduct"
    bottom: "pool5/7x7_s1"
    top: "loss3/classifier"
    param {
	lr_mult: 1
	decay_mult: 1
    }
    param {

joelouismarino

loss3_flat = Flatten()(pool5_7x7_s1)
pool5_drop_7x7_s1 = Dropout(0.4, name='pool5/drop_7x7_s1')(loss3_flat)

joelouismarino

layer {
	name: "pool5/drop_7x7_s1"
	type: "Dropout"
	bottom: "pool5/7x7_s1"
	top: "pool5/7x7_s1"
	dropout_param {
		dropout_ratio: 0.4
	}
}