Hao Wang (王浩)

Why use this restrict? How to do when batch size is small on my machine?

1

https://github.com/ChenRocks/UNITER/blob/80d3602d71d65700eab373acb0507e31e251b7e7/data/sampler.py#L41

How to create word vector for the quickdraw dataset? Is there some tricks to normalize them? like in #10

Would I download all datasets for zero-shot video classification?

The reproduction is very strange

1

When reproduction on Generalized ZSL， the acc of test_unseen is almost 0. I just run your code without any changing. I am very appreciate for your help!

https://github.com/houlu369/Normalized-Quantized-LSTM/blob/8d784679418206343f7a1ca50a6f3d86ffcd0209/model_mnist.py#L158

Recommend one paper using dynamic convolution

2

Context Modulated Dynamic Networks for Actor and Action Video Segmentation with Language Queries， AAAI 2020

The performance is lower than the results in your paper

6

Hi, I tried to reproduce the results on AwA1, without any modification on your code and data. However, I got best_ep: 829, zsl: 0.6889, gzsl: seen=0.7759, unseen=0.5984, h=0.6757 zsl: 0.6889,...