CMC

CMC copied to clipboard

Thanks for open-sourcing your work, I have been trying to use CMC on my custom toy dataset which has 2 views (RGB, Depth). The number of RGB and Depth images...

HaoliangZhou

Pre-trained model for Video task

1

Hi, Could you share the pre-trained model for video action recognition task? i.e. the pre-trained model who's result are shown in Table 2

Shankhanil006

I read the codes and maybe it's for only two views. So, How can it be used for more than two views datasets?

A11en0

Why use learning rate 30~50

3

I saw your note and it seems rather unusual to use such a large learning rate: Note: When training linear classifiers on top of ResNets, it's important to use large...

IgorSusmelj

Hi, Thanks a lot for sharing this great code. In `NCEAverage.py`, the code is shown as the following: ``` if idx is None: idx = self.multinomial.draw(batchSize * (self.K + 1)).view(batchSize,...

LourisXu

when you implement CMC on cifar100, did you resize input image to be 256 by 256, or keep it in its original size

SkyWalker2099

Hi, Thanks for your open-source code. I was training the imagenet on Lab views by nce default setting in your code. I found that the CPU memory is increasing while...

Redaimao

Can anybody tell me how to use the pretrained MoCo?

guomanshan

Hello, thanks for your great work. I find that there is no pre-trained model of ResNet50v1 in DropBox. Could you please provide it?

lzyhha

Hi, Thanks for your excellent work. I have a question about implementation of MoCo. It seems that shuffle_ids from line 440 to line 442 [here](https://github.com/HobbitLong/CMC/blob/7b227be0b10ef4e526c72af07664f5079ed9ee09/train_moco_ins.py#L440) does not work. In other...

Nusselder9