Results 10 comments of Ching-Yao Chuang

Hi, I follow all your instructions. However I try to connect the website using IP/LabelMeAnnotationTool/tool.html and the page stay blank. (IP = xxx.xxx.xx.xx) Do you have any ideas?

I use python 2.7. Btw in `dmn_plus.py`, is it proper to initialize only one gru?

Oops I forget to put them on. Latest commit should fix this issue. Thanks for reporting!

Thank you for correcting my code! You are right, for the experiment I resize the image to 80x80 to reduce observation size.

I've added the __init__.py in the directory, can you try to run the program again?

Thanks for the reply. Do you still have the optimal parameter for learning rate, momentum and decay ratio? Since I follow the learning rate which is 0.05, the loss become...

Thanks for the reply. I use LSTM to encode a sentence, plucking out the hidden state with corresponding sentence length. The first method you mentioned sound reasonable, can I use...

Hi all, I test using mask and padding "all zero" on a visual question answering model and found out that the performance of masking is way better than zero padding....

Sorry for late reply. I construct a mask like 0 0 1 ... 0 0 1 0 ... 0 ... 0 0 0 ... 1 each row only has one...

I have the same problem while running test_tf(). The output of VGG is different from time to time. Any suggestion?