menglin0320

where is the code for roipooling backprop cuda code??

1

As title

opencv error

1

cv2.error: -------src-dir-------/opencv-2.4.10/modules/highgui/src/window.cpp:483: error: (-2) The function is not implemented. where should I do the cmake..

I tried to use python version diff-match-patch to visualize my ocr experiment result. aka difference between gt and prediction. If I just follow the instruction ` dmp = dmp_module.diff_match_patch() diff...

Is there a way to share context among threads, if not why?

10

Basically I want to achieve concurrent work with multithreading and my current inference code is pycuda + tensorrt. **why I want to do so** I'm trying to optimize the inference...

I found that you used elu as the activation function for gan, but on the DCGAN paper they suggest us to use lrelu for the discriminator, relu for encoder and...

question about high level function

1

you did ly.conv2d_transpose(train, 128, 3, stride=2, activation_fn=tf.nn.relu, normalizer_fn=ly.batch_norm, padding='SAME', weights_initializer=tf.random_normal_initializer(0, 0.02)) I wonder if input variable is_training for function ly.batch_norm is always set to False. And if so does batch_norm...

Why you don't have batchnorm for first layer of critic?

2

Some sepcial reason?

Seems you reorgnized the code but didn't add source. to files

2

I don't know why nobody pointed it out, should be a easy fix

I feel that the image_feature should be h4 before activation function and batch normalization. Did you guys try to do it based on the paper writer's code?

sparse_autoencoder_l1, does the l1 constrain really make the representation sparse?

4

Did you find any paper or did you do any empirical experiment that proves that simply adding l1 loss to hidden representation encourages sparsity on the hidden representation