Skip to content

[paligemma] fix processor with suffix #38365

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 1 commit into from
May 27, 2025
+18 −1
Merged

[paligemma] fix processor with suffix #38365

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
5 changes: 4 additions & 1 deletion src/transformers/models/paligemma/processing_paligemma.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,8 @@

from typing import List, Optional, Union

import numpy as np

from ...feature_extraction_utils import BatchFeature
from ...image_utils import ImageInput, is_valid_image, make_flat_list_of_images
from ...processing_utils import (
Expand Down Expand Up @@ -310,7 +312,8 @@ def __call__(
return_data = {**inputs, "pixel_values": pixel_values}

if return_token_type_ids:
labels = inputs["input_ids"].masked_fill(inputs["token_type_ids"] == 0, -100)
labels = np.array(inputs["input_ids"])
labels[np.array(inputs["token_type_ids"]) == 0] = -100
return_data.update({"labels": labels})
return BatchFeature(data=return_data, tensor_type=return_tensors)

Expand Down
14 changes: 14 additions & 0 deletions tests/models/paligemma/test_processor_paligemma.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,6 +62,20 @@ def test_image_seq_length(self):
)
self.assertEqual(len(inputs["input_ids"][0]), 112)

@require_torch
def test_call_with_suffix(self):
input_str = "lower newer"
suffix = "upper older longer string"
image_input = self.prepare_image_inputs()
processor = self.get_processor()
inputs = processor(text=input_str, images=image_input, suffix=suffix)
self.assertTrue("labels" in inputs)
self.assertEqual(len(inputs["labels"][0]), len(inputs["input_ids"][0]))

inputs = processor(text=input_str, images=image_input, suffix=suffix, return_tensors="pt")
self.assertTrue("labels" in inputs)
self.assertEqual(len(inputs["labels"][0]), len(inputs["input_ids"][0]))

def test_text_with_image_tokens(self):
image_processor = self.get_component("image_processor")
tokenizer = self.get_component("tokenizer")
Expand Down