**Note: for network definitions, we directly use `timm.models.ResNet` and [official ConvNeXt](https://github.com/facebookresearch/ConvNeXt/blob/048efcea897d999aed302f2639b6270aedf8d4c8/models/convnext.py).**
Run [/downstream_imagenet/main.py](/downstream_imagenet/main.py) via `torchrun`.
**It is required to specify** the ImageNet data folder (`--data_path`), your experiment name & log dir (`--exp_name` and `--exp_dir`, automatically created if not exists), the model name (`--model`, valid choices see the keys of 'HP_DEFAULT_VALUES' in [/downstream_imagenet/arg.py line14](/downstream_imagenet/arg.py#L14)), and the pretrained weight file `--resume_from` to run fine-tuning.
-`tensorboard_log/`: saves a lot of tensorboard logs, you can visualize accuracies, loss values, learning rates, gradient norms and more things via `tensorboard --logdir /path/to/this/tensorboard_log/ --port 23333`.