-
Ciao ragazzi!
I was wondering whether the new method [AAVAEs](https://arxiv.org/abs/2107.12329) proposed by William Falcon et al. can interest you and thus be implemented in your library or not.
…
-
When finetuning an already trained TransformerBigRelative model with contrastive learning, training loss drops to zero from first iterations and performance degrades due to a higher memory consumption…
-
Hi antoalli,
I have the following error when I tried to train the Deco model using `python train_deco.py --data_root data/shapenetcore_partanno_segmentation_benchmark_v0 --exp_name deco1280_512_wFr…
-
你好,根据文章和issue,测试集应该是对应paper里的table1的50个query,请问它们对应的ground truth在哪呢?
-
- [ ] [Unsupervised Learning by Predicting Noise](https://arxiv.org/abs/1704.05310)
- [ ] [Unsupervised Learning in LSTM Recurrent Neural Networks](http://ftp.idsia.ch/pub/juergen/icann2001unsup.pdf)…
-
Hi; first of all thanks for making this library; it looks very useful.
There are two papers that id like to try and implement in your repo:
[Understanding Contrastive Representation Learning t…
-
- AI News
- [페이스북, 뉴욕대 연구 계정 차단…데이터 연구활용 놓고 갈등](https://n.news.naver.com/mnews/article/001/0012579462?sid=105)
- [DEVIEW 2021 연사 모집](https://deview.kr/2021/cfs)
- 8월 25일까지
- 이번에도 온라인으로…
-
pretrain/contrastive_scene_contexts/lib/shape_context.py Line 89-90
`maskUp = rel_trans[:,:,2] > self.r1`
`maskDown = rel_trans[:,:,2] > self.r1`
maskUp and maskDown are same. Is there existi…
-
code snippet of models.py( line 121-122):
` PGC_labels = torch.zeros([batch_size, 1 + self.queue_size*self.class_num]).cuda()
PGC_labels[:,0:self.queue_size+1].fill_(1.0/(self.queue_s…
-
This is nice work by you and your team.
This is not an issue but rather a question. I am considering combining your ASL loss function with a contrastive loss for the purpose of semi-supervised, m…