Error while converting

[younkyul]

I got an error while converting as follows:

....
I0107 15:12:23.722321 7166 net.cpp:242] This network produces output dense_2
I0107 15:12:23.722332 7166 net.cpp:255] Network initialization done.
Traceback (most recent call last):
File "/home/sr5/younkyu.lee/keras2caffe/keras2caffe-master/convert_youn.py", line 91, in
keras2caffe.convert(keras_model, 'youn.prototxt', 'youn.caffemodel')
File "/home/sr5/younkyu.lee/keras2caffe/keras2caffe-master/keras2caffe/convert.py", line 399, in convert
caffe_model.params[layer][n].data[...] = net_params[layer][n]
ValueError: could not broadcast input array from shape (512,28800) into shape (512,1920)

Process finished with exit code 1

Please help me resolve this issue..

[uhfband]

It's hard to help without knowing of your keras model definition

[younkyul]

Here is the model I used:

model = Sequential()
model.add(Convolution2D(64, (3, 3), padding='same', input_shape=X_train.shape[1:]))
model.add(Activation('relu'))
model.add(Convolution2D(64, (3, 3)), padding='same'))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2), padding='same'))
model.add(Dropout(0.25))

model.add(Convolution2D(128, (3, 3)padding='same'))
model.add(Activation('relu'))
model.add(Convolution2D(128, (3, 3)), padding='same'))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2), padding='same'))
model.add(Dropout(0.25))

model.add(Flatten())
model.add(Dense(512))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(nb_classes))
model.add(Activation('softmax'))

model.compile(loss='binary_crossentropy', optimizer='adadelta', metrics=['accuracy'])

[uhfband]

What is nb_classes and input shape? I will try to reproduce this error.

[younkyul]

Thank you for your help.

• nb_classes: 2
• input shape: (60, 60, 1)

[younkyul]

Would it be because of the version of tools?
I've used:

• Keras 2.2.4
• Tensorflow 1.9.0
• Caffe 1.0

[uhfband]

I had no problem converting your model with Keras 2.1.5. You should try it

[younkyul]

I have tried mine with 2.1.5 and still got the same error.
Could you please share the code for building the model?

[uhfband]

import sys
sys.path.append('../../')

import keras2caffe

from keras.models import Sequential.
from keras.layers import Convolution2D, Activation, Dropout, Dense, Flatten, MaxPooling2D

nb_classes=2

model = Sequential()
#model.add(Convolution2D(64, (3, 3), padding='same', input_shape=X_train.shape[1:]))
model.add(Convolution2D(64, (3, 3), padding='same', input_shape=(60, 60, 1)))
model.add(Activation('relu'))
model.add(Convolution2D(64, (3, 3), padding='same'))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2), padding='same'))
model.add(Dropout(0.25))

model.add(Convolution2D(128, (3, 3), padding='same'))
model.add(Activation('relu'))
model.add(Convolution2D(128, (3, 3), padding='same'))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2), padding='same'))
model.add(Dropout(0.25))

model.add(Flatten())
model.add(Dense(512))
model.add(Activation('relu'))
model.add(Dropout(0.5))
model.add(Dense(nb_classes))
model.add(Activation('softmax'))

model.compile(loss='binary_crossentropy', optimizer='adadelta', metrics=['accuracy'])

keras2caffe.convert(model, 'deploy.prototxt', 'weights.caffemodel')
`

[huangeason26]

i have the same problem @younkyul , and i find that if the "MaxPooling2D" contains "padding='same'" , it can convert well, but it can not work without "padding='same'"