StephenKyung comments

Repositories
Issues
Comments

Results 3 comments of


                                            StephenKyung

onmt_train an illegal memory access was encountered

same here

Why using lua instead of python?

https://github.com/harvardnlp/im2markup/issues/26#issuecomment-558918180 and for one another. Do you think if I changed python code to convert images to greyscare first, can make any improvement?

ValueError: Depth of input (32) is not a multiple of input depth of filter (64) for 'conv2/Conv2D' (op: 'Conv2D') with input shapes: [?,?,?,32], [3,3,64,128].

in tflib\ops.py lin3 238 return tf.nn.max_pool(l_input, ksize=[1, 1, k1, k2], strides=[1, 1, s1, s2], padding='SAME', name=name, data_format='NCHW') should be changed return tf.nn.max_pool(l_input, ksize=[1, k1, k2, 1], strides=[1, s1, s2, 1],...