search

moabitcoin / ig65m-pytorch

PyTorch 3D video classification models pre-trained on 65 million Instagram videos
MIT License
265 stars 30 forks source link

Match the num_class in PyTorch Hub with Models #30

Closed YunseokJANG closed 4 years ago

YunseokJANG commented 4 years ago

[TL;DR] I simply changed the num_class in the PyTorch Hub example to match with the model description.

First of all, thanks for this great work. While I read the Readme file, I found one logical error. So, I made a pull request to correct this, if my understanding is correct.

Specifically, as specified in line #109, "Models fine-tuned on Kinetics have 400 classes, the plain IG65 models 487 (32 clips), and 359 (8 clips) classes", but the PyTorch Hub example sets "359" classes even for the "r2plus1d_34_32_ig65m" model.

Please check the details below.

daniel-j-h commented 4 years ago

Ooops! Thanks for your contributions! :hugs: