improve(utils): add option to load model checkpoint from BytesIO (#1873)

This commit is contained in:
Juan Coria authored and GitHub committed 2025-05-21 14:23:46 +02:00
1 parent 5e56f06866
commit 799fffd768
2 files changed
+18 -7

No files matched your search

+1
View File
@@ -65,6 +65,7 @@ Models can now be stored alongside their pipelines in the same repository, strea
- improve(setup): switch to `lightning` from `pytorch-lightning`
- improve(utils): improve dependency check when loading pretrained models and/or pipeline
- improve(utils): add option to skip dependency check
- improve(utils): add option to load a pretrained model checkpoint from an `io.BytesIO` buffer
### Fixes
+17 -7
View File
@@ -24,6 +24,7 @@
from __future__ import annotations
import os
import io
import warnings
from dataclasses import dataclass
from functools import cached_property
@@ -499,7 +500,7 @@ class Model(lightning.LightningModule):
@classmethod
def from_pretrained(
cls,
checkpoint: Union[Path, Text],
checkpoint: Union[Path, Text, io.BytesIO],
map_location=None,
strict: bool = True,
subfolder: Optional[str] = None,
@@ -512,11 +513,12 @@ class Model(lightning.LightningModule):
Parameters
----------
checkpoint : Path or str
checkpoint : Path, str, or byte buffer
Model checkpoint, provided as one of the following:
* path to a local `pytorch_model.bin` model checkpoint
* path to a local directory containing such a file
* identifier of a model on Huggingface hub
* pre-loaded model checkpoint as a byte buffer
map_location: optional
Same role as in torch.load().
Defaults to `lambda storage, loc: storage`.
@@ -546,9 +548,13 @@ class Model(lightning.LightningModule):
`huggingface_hub.hf_hub_download`
"""
if isinstance(checkpoint, io.BytesIO) or os.path.isfile(checkpoint):
# if checkpoint is a BytesIO object or a file path, use it as is
path_to_model_checkpoint = checkpoint
# if checkpoint is a directory, look for the model checkpoint
# inside this directory (or inside a subfolder if specified)
if os.path.isdir(checkpoint):
elif os.path.isdir(checkpoint):
if subfolder:
path_to_model_checkpoint = (
Path(checkpoint) / subfolder / AssetFileName.Model.value
@@ -556,10 +562,6 @@ class Model(lightning.LightningModule):
else:
path_to_model_checkpoint = Path(checkpoint) / AssetFileName.Model.value
# if checkpoint is a file, use it as is
elif os.path.isfile(checkpoint):
path_to_model_checkpoint = checkpoint
# otherwise, assume that the checkpoint is hosted on HF model hub
else:
_, _, path_to_model_checkpoint = download_from_hf_hub(
@@ -594,6 +596,10 @@ class Model(lightning.LightningModule):
Klass = getattr(module, class_name)
try:
# if checkpoint is a BytesIO object, seek to the beginning so we can load it again
if isinstance(path_to_model_checkpoint, io.BytesIO):
path_to_model_checkpoint.seek(0)
model = Klass.load_from_checkpoint(
path_to_model_checkpoint,
map_location=map_location,
@@ -602,6 +608,10 @@ class Model(lightning.LightningModule):
)
except RuntimeError as e:
if "loss_func" in str(e):
# if checkpoint is a BytesIO object, seek to the beginning so we can load it again
if isinstance(path_to_model_checkpoint, io.BytesIO):
path_to_model_checkpoint.seek(0)
msg = (
"Model has been trained with a task-dependent loss function. "
"Set 'strict' to False to load the model without its loss function "