arxyzan/data2vec-pytorch
PyTorch implementation of "data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language" from Meta AI
PythonMIT
Issues
- 4
In data2vec.py
#20 opened by HarshavardhanaTG - 1
Question on personalized data modality
#19 opened by christincha - 2
Question on the visual modality mask method
#18 opened by Ruizhuo-Xu - 5
Error while training NLP
#17 opened by swarada96 - 1
file type error omegaconf
#15 opened by RuchaCB - 1
Issue with the dependencies
#16 opened by swarada96 - 2
Mask value overflowed in audio pre-training
#14 opened by LuJunru - 3
Some Questions
#13 opened by bryanwong17 - 2
EMA teacher model should not be deepcopied
#12 opened by sudhakaranjain - 7
RuntimeError: Only Tensors created explicitly by the user (graph leaves) support the deepcopy protocol at the moment
#9 opened by Sreyan88 - 1
EMA model forward
#11 opened by anhvth - 3
Any plan for supporting distributed training?
#10 opened by ruizewang - 1
question about disabling gradient bp
#8 opened by XiaoyuShi97 - 2
Question about reproducibility
#7 opened by daisukelab - 2
Trouble with audio....?
#6 opened by drscotthawley - 2
- 2
Model weights?
#4 opened by arxyzan - 1
doubt about finetuning
#3 opened by rafaelvp-db - 2
encoder checkpoint?
#2 opened by kabouzeid