SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Add Idefics2/3 and SmolVLM Fast image processors + improvements for fast image processors (#38157)

Browse Source 
* add working idefics2 fast and improvements for fast nested images processing

* add fast image processors idefics 3 and smolvlm

* cleanup tests

* fic doc idefics2

* PR review and fix issues after merge

* Force providing disable_grouping to group_images_by_shape

* simplify group_images_by_shape

* fix modular

* Fix nits after review
This commit is contained in:
Yoni Gozlan
2025-06-23 10:17:25 -04:00
committed by GitHub
parent 1a96127e46
commit d29482cc91
61 changed files with 2023 additions and 425 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
tests/models/bridgetower/test_image_processing_bridgetower.py
View File

@@ -19,14 +19,11 @@ from typing import Optional, Union
import requests
from transformers.testing_utils import require_torch, require_vision
from transformers.utils import is_torch_available, is_torchvision_available, is_vision_available
from transformers.utils import is_torchvision_available, is_vision_available
from ...test_image_processing_common import ImageProcessingTestMixin, prepare_image_inputs
if is_torch_available():
import torch
if is_vision_available():
from PIL import Image
@@ -124,10 +121,6 @@ class BridgeTowerImageProcessingTest(ImageProcessingTestMixin, unittest.TestCase
self.assertTrue(hasattr(image_processing, "size"))
self.assertTrue(hasattr(image_processing, "size_divisor"))
def _assertEquivalence(self, a, b):
self.assertTrue(torch.allclose(a, b, atol=1e-1))
self.assertLessEqual(torch.mean(torch.abs(a - b)).item(), 1e-3)
@require_vision
@require_torch
def test_slow_fast_equivalence(self):
@@ -146,8 +139,8 @@ class BridgeTowerImageProcessingTest(ImageProcessingTestMixin, unittest.TestCase
encoding_slow = image_processor_slow(dummy_image, return_tensors="pt")
encoding_fast = image_processor_fast(dummy_image, return_tensors="pt")
self._assertEquivalence(encoding_slow.pixel_values, encoding_fast.pixel_values)
self._assertEquivalence(encoding_slow.pixel_mask.float(), encoding_fast.pixel_mask.float())
self._assert_slow_fast_tensors_equivalence(encoding_slow.pixel_values, encoding_fast.pixel_values)
self._assert_slow_fast_tensors_equivalence(encoding_slow.pixel_mask.float(), encoding_fast.pixel_mask.float())
@require_vision
@require_torch
@@ -170,5 +163,5 @@ class BridgeTowerImageProcessingTest(ImageProcessingTestMixin, unittest.TestCase
encoding_slow = image_processor_slow(dummy_images, return_tensors="pt")
encoding_fast = image_processor_fast(dummy_images, return_tensors="pt")
self._assertEquivalence(encoding_slow.pixel_values, encoding_fast.pixel_values)
self._assertEquivalence(encoding_slow.pixel_mask.float(), encoding_fast.pixel_mask.float())
self._assert_slow_fast_tensors_equivalence(encoding_slow.pixel_values, encoding_fast.pixel_values)
self._assert_slow_fast_tensors_equivalence(encoding_slow.pixel_mask.float(), encoding_fast.pixel_mask.float())