seems something worng with "python -m compressai.utils.eval_model --help "

[DreamyEnd]

Bug

when i follow Tutorials-Training-Model evaluation

https://interdigitalinc.github.io/CompressAI/tutorials/tutorial_train.html#model-evaluation

it says "You can run python -m compressai.utils.eval_model --help to get the complete list of options."

To Reproduce

run python -m compressai.utils.eval_model --help

only

usage: __main__.py [-h] {pretrained,checkpoint} ...

Evaluate a model on an image dataset.

positional arguments:
  {pretrained,checkpoint}
                        model source

optional arguments:
  -h, --help            show this help message and exit

Expected behavior

it should be additional information here,right?

Environment

compressai 1.2.6 from pip

PyTorch version: 2.0.1
CUDA used to build PyTorch: 11.7
OS: Ubuntu 20.04.6 LTS (x86_64)
GCC version: (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0
CMake version: version 3.29.2
Libc version: glibc-2.35
Python version: 3.9.19 (main, May  6 2024, 19:43:03)  [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-5.15.0-88-generic-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 11.7.99
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: 
GPU 0: NVIDIA GeForce RTX 3090
GPU 1: NVIDIA GeForce RTX 3090

Nvidia driver version: 525.147.05
cuDNN version: Probably one of the following:
/usr/local/cuda-11.7/targets/x86_64-linux/lib/libcudnn.so.8
/usr/local/cuda-11.7/targets/x86_64-linux/lib/libcudnn_adv_infer.so.8
/usr/local/cuda-11.7/targets/x86_64-linux/lib/libcudnn_adv_train.so.8
/usr/local/cuda-11.7/targets/x86_64-linux/lib/libcudnn_cnn_infer.so.8
/usr/local/cuda-11.7/targets/x86_64-linux/lib/libcudnn_cnn_train.so.8
/usr/local/cuda-11.7/targets/x86_64-linux/lib/libcudnn_ops_infer.so.8
/usr/local/cuda-11.7/targets/x86_64-linux/lib/libcudnn_ops_train.so.8
Is XNNPACK available: True

compressai                1.2.6                  from pip

[YodaEmbedding]

Specify the source:

python -m compressai.utils.eval_model pretrained --help

python -m compressai.utils.eval_model checkpoint --help

However, we should probably make this clearer in the default help text (python -m compressai.utils.eval_model --help), as well as the docs.

[fracape]

Hi, this is the help coming from the parser. For instance, if you use pretrained models, you can run: python -m compressai.utils.eval_model pretained --help to get the help for pretrained models that we provide. (same with checkpoint if you provide your own trained model)