zhengziqiang

I have the same problem with you, have you solved this problem? @vjtrost88

I solved this problem, and I used the wrong img_celeba.zip. It should been downloaded from https://drive.google.com/drive/folders/0B7EVK8r0v71peklHb0pGdDl6R28

Where can I download the sample emoji images?

Will upload the pre-trained weight as soon as possible.

Can you send me the file of deploy.prototxt? When I try to write it , it always wrong. I need your help.Thanks.

ln -s /caffe/build/lib/libcaffe.so /usr/lib/libcaffe.so It works for me

I get the error when I run test_tf(). Is it caused by the version of tf? File "zzq_test.py", line 387, in test_tf fc7 = sess.run(graph.get_tensor_by_name("import/fc7_relu:0"), feed_dict={images:image_val}) File "/home/zzq/.local/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 2561,...

I have the same error, and I found the recurrent_param in the cafffe.proto. So I solved this problem by changing the captioner.py

Can I use gpu for training? I have replaced /cpu:0 with /gpu:0. But it doesn't work.

@ajkl Have you solved this? If yes, pls give me some advice. Thank you