Variable seems to be in the wrong place

[bambidz]

lightning-maml/src/pl/model.py

Line 140 in ce7eedc

test_preds = torch.softmax(train_logit, dim=-1)

According to context, it seems to be test_logit instead of train_logit.

[HansJack1999]

Title:

Error Instantiating pl.LightningDataModule with hydra.utils.instantiate in run.py

Description:

Issue Overview:
No changes were made to the code. When attempting to instantiate a pl.LightningDataModule in run.py using hydra.utils. instantiate, the process fails with an InstantiationException related to the torchmeta.datasets.omniglot.Omniglot class. It appears that the num_classes_per_task parameter is unexpectedly None.

Errror:
The instantiation process throws an InstantiationException, indicating a problem with the num_classes_per_task parameter being None:

发生异常: InstantiationException       (note: full exception trace is shown but execution is paused at: _run_module_as_main)
Error in call to target 'torchmeta.datasets.omniglot.Omniglot':
TypeError("Unknown type for `num_classes_per_task`. Expected `int`, got `<class 'NoneType'>`.")
full_key: data.datamodule.datasets.train
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 92, in _call_target
    return _target_(*args, **kwargs)
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/torchmeta-1.8.0-py3.8.egg/torchmeta/datasets/omniglot.py", line 103, in __init__
    super(Omniglot, self).__init__(dataset, num_classes_per_task,
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/torchmeta-1.8.0-py3.8.egg/torchmeta/utils/data/dataset.py", line 245, in __init__
    raise TypeError('Unknown type for `num_classes_per_task`. Expected '
TypeError: Unknown type for `num_classes_per_task`. Expected `int`, got `<class 'NoneType'>`.

The above exception was the direct cause of the following exception:

  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 97, in _call_target
    raise InstantiationException(msg) from e
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 347, in instantiate_node
    return _call_target(_target_, partial, args, kwargs, full_key)
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 366, in instantiate_node
    cfg[key] = instantiate_node(
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 342, in instantiate_node
    value = instantiate_node(
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/instantiate/_instantiate2.py", line 226, in instantiate
    return instantiate_node(
  File "/home/Code/Few-Shot/lightning-maml-main/src/run.py", line 96, in run
    datamodule: pl.LightningDataModule = hydra.utils.instantiate(
  File "/home/Code/Few-Shot/lightning-maml-main/src/run.py", line 158, in main
    run(cfg)
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/core/utils.py", line 186, in run_job
    ret.return_value = task_function(task_cfg)
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/core/utils.py", line 260, in return_value
    raise self._return_value
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/hydra.py", line 132, in run
    _ = ret.return_value
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/utils.py", line 458, in <lambda>
    lambda: hydra.run(
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/utils.py", line 223, in run_and_report
    raise ex
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/utils.py", line 223, in run_and_report
    raise ex
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/utils.py", line 457, in _run_app
    run_and_report(
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/_internal/utils.py", line 394, in _run_hydra
    _run_app(
  File "/root/anaconda3/envs/metawifi/lib/python3.8/site-packages/hydra/main.py", line 94, in decorated_main
    _run_hydra(
  File "/home/Code/Few-Shot/lightning-maml-main/src/run.py", line 162, in <module>
    main()
  File "/root/anaconda3/envs/metawifi/lib/python3.8/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/root/anaconda3/envs/metawifi/lib/python3.8/runpy.py", line 194, in _run_module_as_main (Current frame)
    return _run_code(code, main_globals, None,
hydra.errors.InstantiationException: Error in call to target 'torchmeta.datasets.omniglot.Omniglot':
TypeError("Unknown type for `num_classes_per_task`. Expected `int`, got `<class 'NoneType'>`.")
full_key: data.datamodule.datasets.train

config：
No changes were made to the configuration

# @package _group_


# The module is communicating with pytorch-lightning
datamodule:
  _target_: pl.datamodule.MetaDataModule


  datasets:
    train:
      _target_: torchmeta.datasets.Omniglot
      root: /home/Code/Few-Shot/lightning-maml-main/data
      meta_train: True
      download: True

    val:
      _target_: torchmeta.datasets.Omniglot
      root: /home/Code/Few-Shot/lightning-maml-main/data
      meta_val: True
      download: True

    test:
      _target_: torchmeta.datasets.Omniglot
      root: /home/Code/Few-Shot/lightning-maml-main/data
      meta_test: True
      download: True

  # number of classes for each task
  nway: 5
  # number of inner steps for adaptation
  num_inner_steps: 1

  # input transformations
  transforms:
    - _target_: torchvision.transforms.Resize
      size: 28
    - _target_: torchvision.transforms.ToTensor


  target_transform:
    _target_: torchmeta.transforms.Categorical

  # this augmentation creates new classes based on rotation.
  class_augmentations:
    - _target_: common.transform.Rotation
      angle:
        - 90
        - 180
        - 270
  # number of tasks for each metadataset split
  batch_size:
    train: 20
    val: 20
    test: 20

  # number of samples per class for a task
  kshot:
    support: 5
    query: 5

  # number of workers to load data to reduce bottleneck
  num_workers:
    train: 2
    val: 2
    test: 2

Environment:
python：3.8.19
pytorch-lightning：2.2.1
torchmet： 1.8.0
hydra-core：1.3.2

Attempts to Resolve:
Checked and verified the configuration files for any obvious misconfigurations.
Searched for similar issues in the repository's issue tracker and on Stack Overflow.
I would appreciate any guidance on resolving this issue to proceed with using the project. Thank you in advance for your help.