Incorrect `.hparams` when using `LightningCLI` with `Callable`

[adosar]

Bug description

The .hparams attribute of a Lightning(Data)Module should store the hyperparameters passed for its initialization. However, when using the LightningCLI it seems that it stores a dict (config-like) object instead of the instantiated object.

What version are you seeing the problem on?

v2.5

How to reproduce the bug

# litmodel.py
import torch
import lightning as L
from collections.abc import Callable


class Model(L.LightningModule):
    def __init__(self, mycallable: Callable):
        super().__init__()
        self.save_hyperparameters()

        print(self.hparams)
        exit()

    def traininig_step(self, batch, batch_idx):
        pass
    
    def configure_optimizers(self):
        pass


if __name__ == '__main__':
    Model(torch.nn.Identity())

# main.py
from lightning.pytorch.cli import LightningCLI
from litmodel import Model
from lightning.pytorch.demos.boring_classes import BoringDataModule


def cli_main():
    return LightningCLI(Model, BoringDataModule)


if __name__ == '__main__':
    cli_main()

# config.yaml
model:
  mycallable:
    class_path: torch.nn.Identity

Error messages and logs

$ python litmodel.py 
/home/asarikas/venvir/common/lib/python3.12/site-packages/lightning/pytorch/utilities/parsing.py:209: Attribute 'mycallable' is an instance of `nn.Module` and is already saved during checkpointing. It is recommended to ignore them using `self.save_hyperparameters(ignore=['mycallable'])`.
"mycallable": Identity()

$ python main.py fit -c config.yaml
/home/asarikas/venvir/common/lib/python3.12/site-packages/lightning/fabric/utilities/seed.py:42: No seed found, seed set to 0
Seed set to 0
"_instantiator": lightning.pytorch.cli.instantiate_module
"mycallable":    {'class_path': 'torch.nn.Identity'}

Environment

/home/asarikas/temp/lightning_cli_bug/collect_env_details.py:24: DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html
import pkg_resources

Current environment

• CUDA:
  • GPU:
    • NVIDIA GeForce RTX 4090
  • available: True
  • version: 12.4
• Lightning:
  • lightning: 2.5.0.post0
  • lightning-utilities: 0.11.9
  • pytorch-lightning: 2.5.0.post0
  • pytorch3d: 0.7.8
  • torch: 2.5.1
  • torch-geometric: 2.6.1
  • torchmetrics: 1.6.1
  • torchvision: 0.20.1
• Packages:
  • absl-py: 2.1.0
  • aenum: 3.1.15
  • aidsorb: 1.0.1.dev59+g0187fc7
  • aiohappyeyeballs: 2.4.4
  • aiohttp: 3.11.11
  • aiosignal: 1.3.2
  • annotated-types: 0.7.0
  • antlr4-python3-runtime: 4.9.3
  • anyio: 4.7.0
  • argon2-cffi: 23.1.0
  • argon2-cffi-bindings: 21.2.0
  • arrow: 1.3.0
  • ase: 3.23.0
  • astroid: 3.3.8
  • asttokens: 3.0.0
  • async-lru: 2.0.4
  • attrs: 24.3.0
  • autocommand: 2.2.2
  • babel: 2.16.0
  • backports.tarfile: 1.2.0
  • beautifulsoup4: 4.12.3
  • bleach: 6.2.0
  • bokeh-sampledata: 2024.2
  • certifi: 2024.12.14
  • cffi: 1.17.1
  • charset-normalizer: 3.4.1
  • comm: 0.2.2
  • contourpy: 1.3.1
  • cycler: 0.12.1
  • debugpy: 1.8.11
  • decorator: 5.1.1
  • defusedxml: 0.7.1
  • dill: 0.3.9
  • docstring-parser: 0.16
  • et-xmlfile: 2.0.0
  • executing: 2.1.0
  • fastjsonschema: 2.21.1
  • filelock: 3.16.1
  • fire: 0.7.0
  • fonttools: 4.55.3
  • fqdn: 1.5.1
  • frozenlist: 1.5.0
  • fsspec: 2024.12.0
  • grpcio: 1.70.0
  • h11: 0.14.0
  • httpcore: 1.0.7
  • httpx: 0.28.1
  • hydra-core: 1.3.2
  • icalendar: 6.1.0
  • idna: 3.10
  • importlib-metadata: 8.0.0
  • importlib-resources: 6.5.2
  • inflect: 7.3.1
  • iopath: 0.1.10
  • ipykernel: 6.29.5
  • ipython: 8.31.0
  • ipywidgets: 8.1.5
  • isoduration: 20.11.0
  • isort: 6.0.0
  • jaraco.collections: 5.1.0
  • jaraco.context: 5.3.0
  • jaraco.functools: 4.0.1
  • jaraco.text: 3.12.1
  • jedi: 0.19.2
  • jinja2: 3.1.5
  • joblib: 1.4.2
  • json5: 0.10.0
  • jsonargparse: 4.37.0
  • jsonpointer: 3.0.0
  • jsonschema: 4.23.0
  • jsonschema-specifications: 2024.10.1
  • jupyter-client: 8.6.3
  • jupyter-core: 5.7.2
  • jupyter-events: 0.11.0
  • jupyter-lsp: 2.2.5
  • jupyter-server: 2.15.0
  • jupyter-server-proxy: 4.4.0
  • jupyter-server-terminals: 0.5.3
  • jupyterlab: 4.3.4
  • jupyterlab-pygments: 0.3.0
  • jupyterlab-server: 2.27.3
  • jupyterlab-vim: 4.1.4
  • jupyterlab-widgets: 3.0.13
  • kiwisolver: 1.4.8
  • latexcodec: 3.0.0
  • lightly: 1.5.19
  • lightly-utils: 0.0.2
  • lightning: 2.5.0.post0
  • lightning-utilities: 0.11.9
  • llvmlite: 0.44.0
  • markdown: 3.7
  • markupsafe: 3.0.2
  • matplotlib: 3.10.0
  • matplotlib-inline: 0.1.7
  • mccabe: 0.7.0
  • mistune: 3.1.0
  • mofdb-client: 0.10.0
  • mofx2csv: 0.1.dev4+g07b461f
  • monty: 2025.1.9
  • more-itertools: 10.6.0
  • mpmath: 1.3.0
  • msgpack: 1.1.0
  • multidict: 6.1.0
  • nbclient: 0.10.2
  • nbconvert: 7.16.5
  • nbformat: 5.10.4
  • nest-asyncio: 1.6.0
  • networkx: 3.4.2
  • notebook-shim: 0.2.4
  • numba: 0.61.0
  • numpy: 1.26.4
  • nvidia-cublas-cu12: 12.4.5.8
  • nvidia-cuda-cupti-cu12: 12.4.127
  • nvidia-cuda-nvrtc-cu12: 12.4.127
  • nvidia-cuda-runtime-cu12: 12.4.127
  • nvidia-cudnn-cu12: 9.1.0.70
  • nvidia-cufft-cu12: 11.2.1.3
  • nvidia-curand-cu12: 10.3.5.147
  • nvidia-cusolver-cu12: 11.6.1.9
  • nvidia-cusparse-cu12: 12.3.1.170
  • nvidia-nccl-cu12: 2.21.5
  • nvidia-nvjitlink-cu12: 12.4.127
  • nvidia-nvtx-cu12: 12.4.127
  • omegaconf: 2.3.0
  • openpyxl: 3.1.5
  • overrides: 7.7.0
  • packaging: 24.2
  • palettable: 3.3.3
  • pandas: 2.2.3
  • pandocfilters: 1.5.1
  • parso: 0.8.4
  • pexpect: 4.9.0
  • pillow: 11.1.0
  • pip: 25.0.1
  • platformdirs: 4.3.6
  • plotly: 5.24.1
  • pooch: 1.8.2
  • portalocker: 3.1.1
  • prometheus-client: 0.21.1
  • prompt-toolkit: 3.0.48
  • propcache: 0.2.1
  • protobuf: 5.29.3
  • psutil: 6.1.1
  • ptyprocess: 0.7.0
  • pure-eval: 0.2.3
  • pybtex: 0.24.0
  • pycparser: 2.22
  • pycryptodome: 3.21.0
  • pydantic: 2.10.6
  • pydantic-core: 2.27.2
  • pygments: 2.18.0
  • pylint: 3.3.4
  • pymatgen: 2025.1.9
  • pymatviz: 0.15.0
  • pymoxel: 0.4.0
  • pynndescent: 0.5.13
  • pyparsing: 3.2.1
  • python-dateutil: 2.9.0.post0
  • python-json-logger: 3.2.1
  • pytorch-lightning: 2.5.0.post0
  • pytorch3d: 0.7.8
  • pytz: 2024.2
  • pyvista: 0.44.2
  • pyyaml: 6.0.2
  • pyzmq: 26.2.0
  • referencing: 0.35.1
  • requests: 2.32.3
  • retnext: 0.1.dev1+gef403b0
  • rfc3339-validator: 0.1.4
  • rfc3986-validator: 0.1.1
  • roma: 1.5.1
  • rpds-py: 0.22.3
  • ruamel.yaml: 0.18.10
  • ruamel.yaml.clib: 0.2.12
  • scikit-learn: 1.6.1
  • scipy: 1.15.0
  • scooby: 0.10.0
  • seaborn: 0.13.2
  • send2trash: 1.8.3
  • sentry-sdk: 1.45.1
  • setuptools: 78.0.2
  • simpervisor: 1.0.0
  • six: 1.17.0
  • sniffio: 1.3.1
  • soupsieve: 2.6
  • spglib: 2.5.0
  • stack-data: 0.6.3
  • stream-inflate: 0.0.41
  • stream-unzip: 0.0.99
  • sympy: 1.13.1
  • tabulate: 0.9.0
  • tenacity: 9.0.0
  • tensorboard: 2.19.0
  • tensorboard-data-server: 0.7.2
  • termcolor: 2.5.0
  • terminado: 0.18.1
  • threadpoolctl: 3.5.0
  • tinycss2: 1.4.0
  • tomli: 2.0.1
  • tomlkit: 0.13.2
  • torch: 2.5.1
  • torch-geometric: 2.6.1
  • torchmetrics: 1.6.1
  • torchvision: 0.20.1
  • tornado: 6.4.2
  • tqdm: 4.67.1
  • traitlets: 5.14.3
  • trame: 3.8.1
  • trame-client: 3.6.1
  • trame-server: 3.4.0
  • trame-vtk: 2.8.15
  • trame-vuetify: 2.9.0
  • triton: 3.1.0
  • typeguard: 4.3.0
  • types-python-dateutil: 2.9.0.20241206
  • typeshed-client: 2.7.0
  • typing-extensions: 4.12.2
  • tzdata: 2024.2
  • umap-learn: 0.5.7
  • uncertainties: 3.2.2
  • uri-template: 1.3.0
  • urllib3: 2.3.0
  • vtk: 9.3.1
  • wcwidth: 0.2.13
  • webcolors: 24.11.1
  • webencodings: 0.5.1
  • websocket-client: 1.8.0
  • werkzeug: 3.1.3
  • wheel: 0.45.1
  • widgetsnbextension: 4.0.13
  • wslink: 2.3.3
  • xyzservices: 2024.9.0
  • yarl: 1.18.3
  • zipp: 3.19.2
• System:
  • OS: Linux
  • architecture:
    • 64bit
    • ELF
  • processor:
  • python: 3.12.7
  • release: 6.12.4-arch1-1
  • version: Proposal for help #1 SMP PREEMPT_DYNAMIC Mon, 09 Dec 2024 14:31:57 +0000

More info

No response

cc @mauvilsa

[mauvilsa]

That is the expected behavior because of feature #18105, so not a bug. It is a dict so that saved checkpoints have all of the information required to instantiate everything from scratch. You can disable this feature by subclassing LightningCLI and adding:

    def _add_instantiators(self):
        pass

I think I will create a pull request with a more official way to disable this. Just note that disabling means that, depending on the case load_from_checkpoint might not work correctly.

[mauvilsa]

In pull request #20777 I propose to add an official way to disable #18105. It would be by doing LightningCLI(... load_from_checkpoint_support=False).

[mauvilsa]

#20777 has been merged so I think this issue can be closed.