search

facebookresearch / ijepa

Official codebase for I-JEPA, the Image-based Joint-Embedding Predictive Architecture. First outlined in the CVPR paper, "Self-supervised learning from images with a joint-embedding predictive architecture."
Other
2.83k stars 358 forks source link

I Just download the image1k dataset from huggingface, what should i do next to process the tar.gz file and train this model #44

Open james2v opened 1 year ago

xiangtaowong commented 1 year ago

i have the same question, too. Looking forwarf to your follow-up progress.

james2v commented 1 year ago

xiangtaowong

l load dataset with this and unzip the datafile into this data structure

1692173103636

from datasets import load_dataset

ds=load_dataset('/imagenet-1k',streaming=True)

and then use the command code it provided from the hub. It works finally.

But i cannot processs the training because of the limit computing source.

xiangtaowong commented 1 year ago

Thanks a lot!