Skip to content

Fix Metaclip modular conversion #40660

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Sep 3, 2025
Merged

Fix Metaclip modular conversion #40660

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
34cfd20
Fix Metaclip modular conversion
Rocketknight1 Sep 3, 2025
dc504fd
manually run check_copies
Rocketknight1 Sep 3, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
44 changes: 20 additions & 24 deletions src/transformers/models/metaclip_2/modeling_metaclip_2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -715,16 +715,14 @@ def forward(
Examples:

```python
>>> import torch
>>> from transformers import AutoTokenizer, MetaClip2TextModelWithProjection

>>> model = MetaClip2TextModelWithProjection.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")
>>> tokenizer = AutoTokenizer.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")

>>> inputs = tokenizer(["a photo of a cat", "a photo of a dog"], padding=True, return_tensors="pt")

>>> with torch.inference_mode():
... outputs = model(**inputs)
>>> outputs = model(**inputs)
>>> text_embeds = outputs.text_embeds
```"""

Expand Down Expand Up @@ -887,9 +885,11 @@ def __init__(self, config: MetaClip2Config):
@auto_docstring
def get_text_features(
self,
input_ids: torch.Tensor,
input_ids: Optional[torch.Tensor] = None,
attention_mask: Optional[torch.Tensor] = None,
position_ids: Optional[torch.Tensor] = None,
output_attentions: Optional[bool] = None,
output_hidden_states: Optional[bool] = None,
) -> torch.FloatTensor:
r"""
Returns:
Expand All @@ -899,16 +899,13 @@ def get_text_features(
Examples:

```python
>>> import torch
>>> from transformers import AutoTokenizer, MetaClip2Model

>>> model = MetaClip2Model.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")
>>> tokenizer = AutoTokenizer.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")

>>> inputs = tokenizer(["a photo of a cat", "a photo of a dog"], padding=True, return_tensors="pt")

>>> with torch.inference_mode():
... text_features = model.get_text_features(**inputs)
>>> text_features = model.get_text_features(**inputs)
```"""
text_outputs: BaseModelOutputWithPooling = self.text_model(
input_ids=input_ids,
Expand All @@ -924,7 +921,9 @@ def get_text_features(
@auto_docstring
def get_image_features(
self,
pixel_values: torch.FloatTensor,
pixel_values: Optional[torch.FloatTensor] = None,
output_attentions: Optional[bool] = None,
output_hidden_states: Optional[bool] = None,
interpolate_pos_encoding: bool = False,
) -> torch.FloatTensor:
r"""
Expand All @@ -935,20 +934,19 @@ def get_image_features(
Examples:

```python
>>> import torch
>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, MetaClip2Model
>>> from transformers.image_utils import load_image

>>> model = MetaClip2Model.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")
>>> processor = AutoProcessor.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = load_image(url)
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> inputs = processor(images=image, return_tensors="pt")

>>> with torch.inference_mode():
... image_features = model.get_image_features(**inputs)
>>> image_features = model.get_image_features(**inputs)
```"""
vision_outputs: BaseModelOutputWithPooling = self.vision_model(
pixel_values=pixel_values,
Expand Down Expand Up @@ -980,22 +978,21 @@ def forward(
Examples:

```python
>>> import torch
>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, MetaClip2Model
>>> from transformers.image_utils import load_image

>>> model = MetaClip2Model.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")
>>> processor = AutoProcessor.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = load_image(url)
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> inputs = processor(
... text=["a photo of a cat", "a photo of a dog"], images=image, return_tensors="pt", padding=True
... )

>>> with torch.inference_mode():
... outputs = model(**inputs)
>>> outputs = model(**inputs)
>>> logits_per_image = outputs.logits_per_image # this is the image-text similarity score
>>> probs = logits_per_image.softmax(dim=1) # we can take the softmax to get the label probabilities
```"""
Expand Down Expand Up @@ -1277,20 +1274,19 @@ def forward(
Examples:

```python
>>> import torch
>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, MetaClip2VisionModelWithProjection
>>> from transformers.image_utils import load_image

>>> model = MetaClip2VisionModelWithProjection.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")
>>> processor = AutoProcessor.from_pretrained("facebook/metaclip-2-worldwide-huge-quickgelu")

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> image = load_image(url)
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> inputs = processor(images=image, return_tensors="pt")

>>> with torch.inference_mode():
... outputs = model(**inputs)
>>> outputs = model(**inputs)
>>> image_embeds = outputs.image_embeds
```"""

Expand Down