Unable to set `val_check_interval` to an integer with the Lightning CLI

[fschlatt]

Bug description

Trying to validate a model every n steps by setting val_check_interval to an integer and check_val_every_n_epochs to None is not possible when using the CLI, because the CLI automatically parses all values to floats. The real issue, however, is that importing lightning changes the order of type annotations of modules. More details below.

What version are you seeing the problem on?

master

Reproduced in studio

No response

How to reproduce the bug

The following minimal example should throw the error:

# main.py
from lightning.pytorch.cli import LightningCLI
from lightning.pytorch.demos.boring_classes import BoringDataModule, DemoModel


def cli_main():
    cli = LightningCLI(DemoModel, BoringDataModule)


if __name__ == "__main__":
    cli_main()

python main.py fit --trainer.val_check_interval 1 --trainer.check_val_every_n_epoch null

I've tried to track down the bug and found that importing lightning changes the order of type annotations. jsonargparse uses this order to verify the types in the CLI, leading to the error. The following code snippets demonstrate the problem. I was unable to find out exactly where or why importing lightning changes the order of annotations, though.

The following two scripts illustrate the problem

from typing import Optional, Union

class Foo:
    def __init__(self, int_or_float: Optional[Union[int, float]] = None):
        self.int_or_float = int_or_float


print(Foo.__init__.__annotations__["int_or_float"])
# typing.Union[int, float, NoneType]

from typing import Optional, Union

import lightning


class Foo:
    def __init__(self, int_or_float: Optional[Union[int, float]] = None):
        self.int_or_float = int_or_float


print(Foo.__init__.__annotations__["int_or_float"])
# typing.Union[float, int, NoneType]

Error messages and logs

Seed set to 0
💡 Tip: For seamless cloud uploads and versioning, try installing [litmodels](https://pypi.org/project/litmodels/) to enable LitModelCheckpoint, which syncs automatically with the Lightning model registry.
Traceback (most recent call last):
  File "/mnt/ceph/storage/data-tmp/2025/fschlatt/pruning-col/minimal_2.py", line 11, in <module>
    cli_main()
  File "/mnt/ceph/storage/data-tmp/2025/fschlatt/pruning-col/minimal_2.py", line 6, in cli_main
    cli = LightningCLI(DemoModel, BoringDataModule)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/conda/lib/python3.11/site-packages/lightning/pytorch/cli.py", line 410, in __init__
    self.instantiate_classes()
  File "/opt/conda/lib/python3.11/site-packages/lightning/pytorch/cli.py", line 604, in instantiate_classes
    self.trainer = self.instantiate_trainer()
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/conda/lib/python3.11/site-packages/lightning/pytorch/cli.py", line 618, in instantiate_trainer
    return self._instantiate_trainer(trainer_config, extra_callbacks)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/conda/lib/python3.11/site-packages/lightning/pytorch/cli.py", line 643, in _instantiate_trainer
    return self.trainer_class(**config)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/conda/lib/python3.11/site-packages/lightning/pytorch/utilities/argparse.py", line 70, in insert_env_defaults
    return fn(self, **kwargs)
           ^^^^^^^^^^^^^^^^^^
  File "/opt/conda/lib/python3.11/site-packages/lightning/pytorch/trainer/trainer.py", line 445, in __init__
    self._data_connector.on_trainer_init(
  File "/opt/conda/lib/python3.11/site-packages/lightning/pytorch/trainer/connectors/data_connector.py", line 65, in on_trainer_init
    raise MisconfigurationException(
lightning.fabric.utilities.exceptions.MisconfigurationException: `val_check_interval` should be an integer when `check_val_every_n_epoch=None`, found 1.0.```

### Environment

<details>
  <summary>Current environment</summary>

Current environment

• CUDA:
  • GPU:
    • NVIDIA A100-SXM4-40GB MIG 3g.20gb
  • available: True
  • version: 12.8
• Lightning:
  • lightning: 2.5.5
  • lightning-ir: 0.0.4
  • lightning-utilities: 0.15.2
  • pytorch-lightning: 2.5.5
  • torch: 2.7.1+cu128
  • torchaudio: 2.7.1+cu128
  • torchdata: 0.11.0
  • torchelastic: 0.2.2
  • torchmetrics: 1.8.2
  • torchvision: 0.22.1+cu128
• Packages:
  • accelerate: 1.10.1
  • aiohappyeyeballs: 2.6.1
  • aiohttp: 3.12.15
  • aiosignal: 1.4.0
  • annotated-types: 0.7.0
  • antlr4-python3-runtime: 4.9.3
  • anyio: 4.10.0
  • aquarel: 0.0.7
  • archspec: 0.2.5
  • argon2-cffi: 25.1.0
  • argon2-cffi-bindings: 25.1.0
  • arrow: 1.3.0
  • asttokens: 3.0.0
  • astunparse: 1.6.3
  • async-lru: 2.0.5
  • attrs: 25.3.0
  • autocommand: 2.2.2
  • babel: 2.17.0
  • backports.tarfile: 1.2.0
  • beautifulsoup4: 4.13.4
  • beir: 2.2.0
  • bitarray: 3.7.1
  • black: 25.1.0
  • bleach: 6.2.0
  • blinker: 1.9.0
  • boltons: 24.0.0
  • brotli: 1.1.0
  • bs4: 0.0.2
  • cbor: 1.0.0
  • certifi: 2025.8.3
  • cffi: 1.17.1
  • chardet: 5.2.0
  • charset-normalizer: 3.4.2
  • chest: 0.2.3
  • click: 8.2.1
  • cmake: 4.0.2
  • colbert-ai: 0.2.19
  • colorama: 0.4.6
  • coloredlogs: 15.0.1
  • comm: 0.2.3
  • conda: 25.7.0
  • conda-build: 25.5.0
  • conda-index: 0.6.1
  • conda-libmamba-solver: 25.3.0
  • conda-package-handling: 2.4.0
  • conda-package-streaming: 0.11.0
  • contourpy: 1.3.3
  • cycler: 0.12.1
  • cython: 3.1.4
  • datasets: 4.1.0
  • debugpy: 1.8.16
  • decorator: 5.2.1
  • defusedxml: 0.7.1
  • deprecated: 1.2.18
  • dill: 0.4.0
  • distro: 1.9.0
  • dnspython: 2.7.0
  • docker: 7.1.0
  • docstring-parser: 0.17.0
  • einops: 0.8.1
  • evalidate: 2.0.5
  • exceptiongroup: 1.3.0
  • executing: 2.2.0
  • expecttest: 0.3.0
  • faiss: 1.9.0
  • fastapi: 0.116.2
  • fastchat: 0.1.0
  • fastjsonschema: 2.21.2
  • fastkmeans: 0.5.0
  • filelock: 3.18.0
  • flash-attn: 2.8.3
  • flask: 3.1.2
  • flatbuffers: 25.2.10
  • fonttools: 4.59.2
  • fqdn: 1.5.1
  • frozendict: 2.4.6
  • frozenlist: 1.7.0
  • fsspec: 2025.5.1
  • ftfy: 6.3.1
  • git-python: 1.0.3
  • gitdb: 4.0.12
  • gitpython: 3.1.45
  • h11: 0.16.0
  • h2: 4.2.0
  • heapdict: 1.0.1
  • hf-xet: 1.1.10
  • hpack: 4.1.0
  • httpcore: 1.0.9
  • httpx: 0.28.1
  • httpx-sse: 0.4.1
  • huggingface-hub: 0.35.0
  • humanfriendly: 10.0
  • hyperframe: 6.1.0
  • hypothesis: 6.135.0
  • idna: 3.10
  • ijson: 3.4.0
  • importlib-metadata: 8.0.0
  • importlib-resources: 6.5.2
  • inflect: 7.3.1
  • iniconfig: 2.1.0
  • inscriptis: 2.6.0
  • ipykernel: 6.30.1
  • ipython: 9.3.0
  • ipython-pygments-lexers: 1.1.1
  • ipywidgets: 8.1.7
  • ir-datasets: 0.5.11
  • ir-measures: 0.4.1
  • isoduration: 20.11.0
  • isort: 6.0.1
  • itsdangerous: 2.2.0
  • 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: 3.1.6
  • jiter: 0.11.0
  • joblib: 1.5.2
  • json5: 0.12.1
  • jsonargparse: 4.41.0
  • jsonlines: 4.0.0
  • jsonpatch: 1.33
  • jsonpointer: 3.0.0
  • jsonschema: 4.24.0
  • jsonschema-specifications: 2025.4.1
  • jupyter: 1.1.1
  • jupyter-client: 8.6.3
  • jupyter-console: 6.6.3
  • jupyter-core: 5.8.1
  • jupyter-events: 0.12.0
  • jupyter-lsp: 2.3.0
  • jupyter-server: 2.17.0
  • jupyter-server-terminals: 0.5.3
  • jupyterlab: 4.4.7
  • jupyterlab-pygments: 0.3.0
  • jupyterlab-server: 2.27.3
  • jupyterlab-widgets: 3.0.15
  • kiwisolver: 1.4.9
  • libarchive-c: 5.3
  • libmambapy: 2.1.1
  • lief: 0.16.4
  • lightning: 2.5.5
  • lightning-ir: 0.0.4
  • lightning-utilities: 0.15.2
  • lintrunner: 0.12.7
  • lxml: 6.0.1
  • lz4: 4.4.4
  • markupsafe: 3.0.2
  • matplotlib: 3.10.6
  • matplotlib-inline: 0.1.7
  • maturin: 1.9.4
  • mcp: 1.14.0
  • menuinst: 2.2.0
  • mistune: 3.1.4
  • more-itertools: 10.4.0
  • mpmath: 1.3.0
  • msgpack: 1.1.0
  • multidict: 6.6.4
  • multiprocess: 0.70.16
  • mypy-extensions: 1.1.0
  • nbclient: 0.10.2
  • nbconvert: 7.16.6
  • nbformat: 5.10.4
  • nest-asyncio: 1.6.0
  • networkx: 3.5
  • ninja: 1.11.1.4
  • nltk: 3.9.1
  • notebook: 7.4.5
  • notebook-shim: 0.2.4
  • numpy: 1.26.4
  • nvidia-cublas-cu12: 12.8.3.14
  • nvidia-cuda-cupti-cu12: 12.8.57
  • nvidia-cuda-nvrtc-cu12: 12.8.61
  • nvidia-cuda-runtime-cu12: 12.8.57
  • nvidia-cudnn-cu12: 9.7.1.26
  • nvidia-cufft-cu12: 11.3.3.41
  • nvidia-cufile-cu12: 1.13.0.11
  • nvidia-curand-cu12: 10.3.9.55
  • nvidia-cusolver-cu12: 11.7.2.55
  • nvidia-cusparse-cu12: 12.5.7.53
  • nvidia-cusparselt-cu12: 0.6.3
  • nvidia-nccl-cu12: 2.26.2
  • nvidia-nvjitlink-cu12: 12.8.61
  • nvidia-nvtx-cu12: 12.8.55
  • omegaconf: 2.3.0
  • onnxruntime: 1.22.1
  • openai: 1.107.3
  • optree: 0.16.0
  • overrides: 7.7.0
  • packaging: 25.0
  • pandas: 2.3.2
  • pandocfilters: 1.5.1
  • parso: 0.8.4
  • pathspec: 0.12.1
  • patsy: 1.0.1
  • pexpect: 4.9.0
  • pickleshare: 0.7.5
  • pillow: 11.0.0
  • pip: 25.1.1
  • pkginfo: 1.12.1.2
  • pkgutil-resolve-name: 1.3.10
  • platformdirs: 4.3.8
  • pluggy: 1.5.0
  • prometheus-client: 0.22.1
  • prompt-toolkit: 3.0.51
  • propcache: 0.3.2
  • protobuf: 6.32.1
  • psutil: 7.0.0
  • ptranking: 0.0.5
  • ptyprocess: 0.7.0
  • pure-eval: 0.2.3
  • pyarrow: 21.0.0
  • pycosat: 0.6.6
  • pycparser: 2.22
  • pydantic: 2.11.9
  • pydantic-core: 2.33.2
  • pydantic-settings: 2.10.1
  • pygments: 2.19.1
  • pyjnius: 1.7.0
  • pyndeval: 0.0.6
  • pyparsing: 3.2.4
  • pyserini: 1.2.0
  • pysocks: 1.7.1
  • pytest: 8.4.2
  • python-dateutil: 2.9.0.post0
  • python-dotenv: 1.1.1
  • python-etcd: 0.4.5
  • python-json-logger: 3.3.0
  • python-multipart: 0.0.20
  • python-terrier: 0.13.1
  • pytorch-lightning: 2.5.5
  • pytrec-eval-terrier: 0.5.9
  • pytz: 2025.2
  • pyyaml: 6.0.2
  • pyzmq: 27.1.0
  • referencing: 0.36.2
  • regex: 2025.9.1
  • requests: 2.32.3
  • rfc3339-validator: 0.1.4
  • rfc3986-validator: 0.1.1
  • rpds-py: 0.25.1
  • ruamel-yaml: 0.18.12
  • ruamel.yaml.clib: 0.2.8
  • safetensors: 0.6.2
  • sarge: 0.1.7.post1
  • scikit-learn: 1.7.2
  • scipy: 1.16.2
  • seaborn: 0.13.2
  • send2trash: 1.8.3
  • sentence-transformers: 5.1.0
  • sentencepiece: 0.2.1
  • sentry-sdk: 2.38.0
  • setuptools: 80.9.0
  • six: 1.17.0
  • smmap: 5.0.2
  • sniffio: 1.3.1
  • sortedcontainers: 2.4.0
  • soupsieve: 2.7
  • sse-starlette: 3.0.2
  • stack-data: 0.6.3
  • starlette: 0.48.0
  • statsmodels: 0.14.5
  • sympy: 1.14.0
  • terminado: 0.18.1
  • threadpoolctl: 3.6.0
  • tiktoken: 0.11.0
  • tinycss2: 1.4.0
  • tira: 0.0.178
  • tokenizers: 0.22.0
  • tomli: 2.0.1
  • torch: 2.7.1+cu128
  • torchaudio: 2.7.1+cu128
  • torchdata: 0.11.0
  • torchelastic: 0.2.2
  • torchmetrics: 1.8.2
  • torchvision: 0.22.1+cu128
  • tornado: 6.5.2
  • tqdm: 4.67.1
  • traitlets: 5.14.3
  • transformers: 4.56.1
  • trec-car-tools: 2.6
  • trectools: 0.0.50
  • triton: 3.3.1
  • truststore: 0.10.1
  • typeguard: 4.3.0
  • types-dataclasses: 0.6.6
  • types-python-dateutil: 2.9.0.20250822
  • typeshed-client: 2.8.2
  • typing-extensions: 4.14.0
  • typing-inspection: 0.4.1
  • tzdata: 2025.2
  • ujson: 5.11.0
  • unlzw3: 0.2.3
  • uri-template: 1.3.0
  • urllib3: 2.4.0
  • uvicorn: 0.35.0
  • wandb: 0.21.4
  • warc3-wet: 0.2.5
  • warc3-wet-clueweb09: 0.2.5
  • 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: 4.0.14
  • wrapt: 1.17.3
  • xxhash: 3.5.0
  • yarl: 1.20.1
  • zipp: 3.22.0
  • zlib-state: 0.1.10
  • zstandard: 0.23.0
• System:
  • OS: Linux
  • architecture:
    • 64bit
  • processor: x86_64
  • python: 3.11.13
  • release: 6.11.0-26-generic
  • version: Cannot load saved model. #26~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Apr 17 19:20:47 UTC 2

</details>

cc @mauvilsa

[SkafteNicki]

Hi @fschlatt, thanks for reporting this issue. I am trying to reproduce the issue on my laptop and is unable in both cases. Both the first example with setting val_check_interval in combination with lightning cli works perfectly fine and the training starts as expected and for the second example I do not see a change in the order of the annotation, see screenshot as documentation:

Tested on v2.5.5.

[fschlatt]

Interesting. The bug occurs in this docker image, but I have not tested it locally on my own laptop. I will investigate further ASAP and report back.

[fschlatt]

@SkafteNicki I can reproduce it for python >= 3.12. When creating fresh conda environments and only installing lightning, I get the error for python 3.12 and 3.13, but not for 3.11 (test.py is the example script above)

$ conda activate test-11
$ python --version
Python 3.11.13
$ python test.py 
typing.Union[int, float, NoneType]
$ conda activate test-12
$ python --version
Python 3.12.11
$ python test.py 
typing.Union[float, int, NoneType]
$ conda activate test-13
$ python --version
Python 3.13.7
$ python test.py 
typing.Union[float, int, NoneType]

[tshu-w]

Encountered the same issue.

This also happened on 3.11.13 on my side.