SuiAn comments

Results 11 comments of


                                            SuiAn

ImportError: No module named 'openai_benchmark'

how could i run your code if i use gym in new version?

The atari's state.shape is (210,160,3),and u define the net's in_channel as env.shape[0]

Is there anyone find the same issue or could answer for me, thanks a lot!

The atari's state.shape is (210,160,3),and u define the net's in_channel as env.shape[0]

Thanks for your answer. This really help me solve the problem.

Did lstm cell really make more sense in A3C?

Just now, i have a test for whether using the lstm cell. The result shows when using lstm cell, our net will converge faster than not. In GTX950, the time...

The generated samples are not great for WAE_MMD

> Thank you very much for your reply, I will try your suggested tips.

AAN dataset crashing when loading .tsv file

Hi, there, The provided download URL no longer works now, could anyone share the data, many thx!

AAN dataset crashing when loading .tsv file

> > Hi, there, > > The provided download URL no longer works now, could anyone share the data, many thx! > > Have you figured out any way to...

The function in Class: LocalAggregationLossModule

Hi, a similar question. When I use the _softmax() function, the output prob is always too large to optimize (about 1600), since 2876934.2 does not make sence, can I manually...

The function in Class: LocalAggregationLossModule

Thanks for your reply, the original implementation of [Instance Discrimination](https://github.com/zhirongw/lemniscate.pytorch) is here. I will check it.

About accuracy of baseline in miniImagenet

@ZhangMingliangAI Hi, I just tested on 5-way 5-shot experiments as the default setting. For the baseline, the result is 56.92%. For MatchingNet the result is 59.81%(reported 63.48%) and for ProtoNet,...