Export the model checkpoint to an onnx model. Typical usage examples: .. code-block:: bash python -m monai.bundle onnx_export network --filepath --ckpt_file ... Args: net_id: ID name of the network component in the config, it must
(
net_id: str | None = None,
filepath: PathLike | None = None,
ckpt_file: str | None = None,
meta_file: str | Sequence[str] | None = None,
config_file: str | Sequence[str] | None = None,
key_in_ckpt: str | None = None,
use_trace: bool | None = None,
input_shape: Sequence[int] | None = None,
args_file: str | None = None,
converter_kwargs: Mapping | None = None,
**override: Any,
)
| 1321 | |
| 1322 | |
| 1323 | def onnx_export( |
| 1324 | net_id: str | None = None, |
| 1325 | filepath: PathLike | None = None, |
| 1326 | ckpt_file: str | None = None, |
| 1327 | meta_file: str | Sequence[str] | None = None, |
| 1328 | config_file: str | Sequence[str] | None = None, |
| 1329 | key_in_ckpt: str | None = None, |
| 1330 | use_trace: bool | None = None, |
| 1331 | input_shape: Sequence[int] | None = None, |
| 1332 | args_file: str | None = None, |
| 1333 | converter_kwargs: Mapping | None = None, |
| 1334 | **override: Any, |
| 1335 | ) -> None: |
| 1336 | """ |
| 1337 | Export the model checkpoint to an onnx model. |
| 1338 | |
| 1339 | Typical usage examples: |
| 1340 | |
| 1341 | .. code-block:: bash |
| 1342 | |
| 1343 | python -m monai.bundle onnx_export network --filepath <export path> --ckpt_file <checkpoint path> ... |
| 1344 | |
| 1345 | Args: |
| 1346 | net_id: ID name of the network component in the config, it must be `torch.nn.Module`. |
| 1347 | filepath: filepath where the onnx model is saved to. |
| 1348 | ckpt_file: filepath of the model checkpoint to load. |
| 1349 | meta_file: filepath of the metadata file, if it is a list of file paths, the content of them will be merged. |
| 1350 | config_file: filepath of the config file that contains extract network information, |
| 1351 | key_in_ckpt: for nested checkpoint like `{"model": XXX, "optimizer": XXX, ...}`, specify the key of model |
| 1352 | weights. if not nested checkpoint, no need to set. |
| 1353 | use_trace: whether using `torch.jit.trace` to convert the pytorch model to torchscript model. |
| 1354 | input_shape: a shape used to generate the random input of the network, when converting the model to an |
| 1355 | onnx model. Should be a list like [N, C, H, W] or [N, C, H, W, D]. If not given, will try to parse from |
| 1356 | the `metadata` config. |
| 1357 | args_file: a JSON or YAML file to provide default values for all the parameters of this function, so that |
| 1358 | the command line inputs can be simplified. |
| 1359 | converter_kwargs: extra arguments that are needed by `convert_to_onnx`, except ones that already exist in the |
| 1360 | input parameters. |
| 1361 | override: id-value pairs to override or add the corresponding config content. |
| 1362 | e.g. ``--_meta#network_data_format#inputs#image#num_channels 3``. |
| 1363 | |
| 1364 | """ |
| 1365 | _args = update_kwargs( |
| 1366 | args=args_file, |
| 1367 | net_id=net_id, |
| 1368 | filepath=filepath, |
| 1369 | meta_file=meta_file, |
| 1370 | config_file=config_file, |
| 1371 | ckpt_file=ckpt_file, |
| 1372 | key_in_ckpt=key_in_ckpt, |
| 1373 | use_trace=use_trace, |
| 1374 | input_shape=input_shape, |
| 1375 | converter_kwargs=converter_kwargs, |
| 1376 | **override, |
| 1377 | ) |
| 1378 | _log_input_summary(tag="onnx_export", args=_args) |
| 1379 | ( |
| 1380 | filepath_, |
nothing calls this directly
no test coverage detected
searching dependent graphs…