-
Dear authors, It seems that you still follow a local-global contrastive manner when applying the model to node classification tasks(it's similar to 'mvgrl' but adopts different data augmentations). Is…
-
### Brief Summary:
Describe the what, why, and how of your content idea in 2-5 sentences.
Constrastive Learning date as far back as the 1990s. It is a learning paradigm where a representation is …
-
Hey Yunfan,
first of all, this is really great work and a well written paper! Thanks for providing the code.
I am trying to reimplement your method and am a bit confused about the way you penali…
-
I'm dealing with a weird scenario where I have two different audio denoising models and I want to train them in two specific procedures. I like how the System() class has a `common_step` function, so …
-
Be sure to visit our [Resources Page](https://github.com/section-engineering-education/engineering-education/blob/master/new_contributors/resources-page.md) for tools, resources, and example articles …
-
首先我对您的论文很感兴趣,最近也正在尝试复现,但是在这个过程中我发现了一个很有意思的现象,在使用github的代码复现过程中pre-training改变--neg,不管是repeat或者是shuffle,best model都是best_model_141.pt,接下来fine-tune进行下游识别任务时,ucf101的acc-avg都是73.5,感到很奇怪,是不是哪里出现问题了,…
-
From the original paper,
> For the key encoder f_k, we shuffle the sample order in the current mini-batch before distributing it among GPUs (and shuffle back after encoding); the sample order of the …
-
Hi, In your paper “Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID”, there was one particular place that gets me a little bit confuse.
![image](https://user-image…
-
Hi,
I like the list of papers! Thanks a lot for sharing. One of my recent favorites is missing: https://proceedings.neurips.cc/paper/2020/file/e2ef524fbf3d9fe611d5a8e90fefdc9c-Paper.pdf
They use s…
-
Hi,
Recently, I want to reproduce the results on the imagenet with the config of **src/configs/ILSVRC2012/ContraGAN256.json** . The command is:
> `python src/main.py -t -e -sync_bn -c …