facebookresearch/jepa
PyTorch code and models for V-JEPA self-supervised learning from video.
PythonNOASSERTION
Issues
- 4
- 1
Dimensional Collapse Tracking
#77 opened by zankerx - 0
- 3
is it possible to learn an action model (or the action's effects) with v-jepa ?
#71 opened by aymeric75 - 1
Where are the model weights
#74 opened by DrJessop - 0
Question about Eval code
#73 opened by saunair - 0
- 0
- 1
Is AutoAugment used for videos or not
#65 opened by Separius - 1
Release of "conditional diffusion decoder"
#53 opened by andyshuo - 2
- 0
- 0
PCA feature map visualization of a pre-trained weights look very random, compared to without pre-trained weights loaded
#66 opened by icekang - 0
video_dataset: Random sample assignment
#64 opened by thomasf1 - 0
error when ruuning the evaluation commands
#62 opened by Ahmed-Elgamel - 4
ValueError: Default process group has not been initialized, please make sure to call init_process_group
#55 opened by MKaczkow - 0
Question on number of heads in Attentive Pooler
#60 opened by orrzohar - 1
long video
#44 opened by betterze - 0
data loading during eval
#59 opened by orrzohar - 0
Running evaluation without training
#58 opened by orrzohar - 2
ViT-tiny encoder
#22 opened by jez-moxmo - 0
Any plan of releasing decoder checkpoint?
#54 opened by tiangeluo - 2
FSDP Support
#51 opened by andrew-bydlon - 11
- 2
Question about the mask sampling
#50 opened by FriedRonaldo - 0
- 1
Use of Conditional (possibly latent) variables (z) in predictor from original JEPA paper.
#46 opened by Sharpz7 - 0
Question about expected finetuning performance
#47 opened by VGrondin - 1
KeyError when running evals.main
#43 opened by JPerAsperaadAstra - 7
Crashes after first Epoch
#26 opened by thomasf1 - 1
- 2
Evaluation / Testing data
#33 opened by aakgun - 0
Evals not eval for evaluation script (typo)
#41 opened by atonkamanda - 0
numpy import error
#40 opened by tomarvimal - 1
Question: no release of decoder checkpoint?
#37 opened by LumenYoung - 0
- 2
Cannot access the config file vitl16_k400.yaml
#34 opened by jmgrn61 - 3
Potential bug
#32 opened by dnnspark - 2
Augmentations in the frozen evaluation
#24 opened by bryant1410 - 0
few questions on feature detection
#30 opened by jez-moxmo - 0
problem about '$slurm_partition'
#25 opened by wnzhyee - 0
- 2
Vram required?
#3 opened by Koorvux - 1
ImageNet1K Attentive probes checkpoint links are broken for ViT-L/16 and ViT-H/16
#16 opened by bryant1410 - 0
research
#23 opened by csprinkle74 - 1
- 0
- 1
- 1
Evals, not Eval
#10 opened by JustinYuu - 1
typo/bug in pooler?
#4 opened by N0r9st