run.py issues

[ilhembekkr]

Hello,

I am experiencing persistent issues while attempting to run the code and now particularly the dev_medmnist Docker image. Despite trying various branches and experiments, I consistently encounter errors that prevent successful execution.

Could you please provide a functional Dockerfile along with the correct versions and configurations for running the experiments? Detailed instructions or any troubleshooting tips would also be greatly appreciated.

Thank you for your assistance.

(myenv) root@cce420855731:/workspace/project# python run.py
[2024-07-02 08:03:01,627] [INFO] [/workspace/project/hyperbox/utils/utils.py:101] - Disabling python warnings! <config.ignore_warnings=True>
[2024-07-02 08:03:01,647] [INFO] [/workspace/project/hyperbox/train.py:36] - Instantiating datamodule <hyperbox.datamodules.mnist_datamodule.MNISTDataModule>
[2024-07-02 08:03:02,248][matplotlib.font_manager][INFO] - generated new fontManager
[2024-07-02 08:03:02,414] [INFO] [/workspace/project/hyperbox/train.py:40] - Instantiating model <hyperbox.models.mnist_model.MNISTLitModel>
Error executing job with overrides: []
Traceback (most recent call last):
  File "/opt/conda/envs/myenv/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 62, in _call_target
    return target(*args, **kwargs)
  File "/workspace/project/hyperbox/models/mnist_model.py", line 49, in __init__
    self.train_accuracy = Accuracy()
TypeError: __new__() missing 1 required positional argument: 'task'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "run.py", line 34, in main
    return train(config)
  File "/workspace/project/hyperbox/train.py", line 43, in train
    model: LightningModule = hydra.utils.instantiate(config.model, _recursive_=False)
  File "/opt/conda/envs/myenv/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 180, in instantiate
    return instantiate_node(config, *args, recursive=_recursive_, convert=_convert_)
  File "/opt/conda/envs/myenv/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 249, in instantiate_node
    return _call_target(target, *args, **kwargs)
  File "/opt/conda/envs/myenv/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 64, in _call_target
    raise type(e)(
  File "/opt/conda/envs/myenv/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 62, in _call_target
    return target(*args, **kwargs)
  File "/workspace/project/hyperbox/models/mnist_model.py", line 49, in __init__
    self.train_accuracy = Accuracy()
TypeError: Error instantiating 'hyperbox.models.mnist_model.MNISTLitModel' : __new__() missing 1 required positional argument: 'task'

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.

I have tried using different commands and specifying various experiment names, **even with fixing the Accuracy(task="multiclass', mum_classes=10) **I always had other errors, and nothing seems to work. Could you please provide a functional Dockerfile along with the correct versions and configurations for running the experiments? and an example of an execution scenario that works would be greatly appreciated.