riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-08-02 03:01:07 +06:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

again

Browse Source
This commit is contained in:
Quentin Lhoest 2025-06-20 13:19:50 +02:00
parent 3dfebf2fc0
commit 1fdb9f3908
4 changed files with 9 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
tests/models/perceiver/test_modeling_perceiver.py
View File

@ -842,11 +842,8 @@ def prepare_img():
# Helper functions for optical flow integration test
def prepare_optical_flow_images():
dataset = load_dataset("hf-internal-testing/fixtures_sintel", split="test")
image1 = Image.open(dataset[0]["file"]).convert("RGB")
image2 = Image.open(dataset[0]["file"]).convert("RGB")
return image1, image2
ds = load_dataset("hf-internal-testing/fixtures_sintel", split="test")
return list(ds["image"][:2])
def normalize(img):

6
tests/models/vilt/test_modeling_vilt.py
View File

@ -637,9 +637,9 @@ class ViltModelIntegrationTest(unittest.TestCase):
processor = self.default_processor
dataset = load_dataset("hf-internal-testing/fixtures_nlvr2", split="test")
image1 = Image.open(dataset[0]["file"]).convert("RGB")
image2 = Image.open(dataset[1]["file"]).convert("RGB")
dataset = load_dataset("hf-internal-testing/fixtures_nlvr2", split="train")
image1 = dataset[0]["image"]
image2 = dataset[1]["image"]
text = (
"The left image contains twice the number of dogs as the right image, and at least two dogs in total are"

6
tests/models/vision_encoder_decoder/test_modeling_vision_encoder_decoder.py
View File

@ -1149,8 +1149,8 @@ class TrOCRModelIntegrationTest(unittest.TestCase):
def test_inference_handwritten(self):
model = VisionEncoderDecoderModel.from_pretrained("microsoft/trocr-base-handwritten").to(torch_device)
dataset = load_dataset("hf-internal-testing/fixtures_ocr", split="test")
image = Image.open(dataset[0]["file"]).convert("RGB")
dataset = load_dataset("hf-internal-testing/fixtures_ocr", split="train")
image = dataset[0]["image"]
processor = self.default_processor
pixel_values = processor(images=image, return_tensors="pt").pixel_values.to(torch_device)
@ -1175,7 +1175,7 @@ class TrOCRModelIntegrationTest(unittest.TestCase):
model = VisionEncoderDecoderModel.from_pretrained("microsoft/trocr-base-printed").to(torch_device)
dataset = load_dataset("hf-internal-testing/fixtures_ocr", split="test")
image = Image.open(dataset[1]["file"]).convert("RGB")
image = dataset[0]["image"]
processor = self.default_processor
pixel_values = processor(images=image, return_tensors="pt").pixel_values.to(torch_device)

4
tests/pipelines/test_pipelines_automatic_speech_recognition.py
View File

@ -265,9 +265,7 @@ class AutomaticSpeechRecognitionPipelineTests(unittest.TestCase):
@require_torch
@require_pyctcdecode
def test_large_model_pt_with_lm(self):
dataset = load_dataset("Narsil/asr_dummy", streaming=True)
third_item = next(iter(dataset["test"].skip(3)))
filename = third_item["file"]
filename = hf_hub_download("Narsil/asr_dummy", filename="4.flac", repo_type="dataset")
speech_recognizer = pipeline(
task="automatic-speech-recognition",