dnrops
/
transformers
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix all torch pipeline failures except one (#30290) 

* fix

* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
This commit is contained in:
Yih-Dar 2024-04-18 10:35:43 +02:00 committed by GitHub
parent 7915a25976
commit 28a22834bf
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 25 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/transformers/pipelines/__init__.py
View File

@ -367,8 +367,8 @@ SUPPORTED_TASKS = {
"pt": (AutoModel,) if is_torch_available() else (),
"default": {
"model": {
"pt": ("google/vit-base-patch16-224", "29e7a1e183"),
"tf": ("google/vit-base-patch16-224", "29e7a1e183"),
"pt": ("google/vit-base-patch16-224", "3f49326"),
"tf": ("google/vit-base-patch16-224", "3f49326"),
}
},
"type": "image",

2
tests/pipelines/test_pipelines_automatic_speech_recognition.py
View File

@ -1216,7 +1216,7 @@ class AutomaticSpeechRecognitionPipelineTests(unittest.TestCase):
@slow
def test_whisper_longform(self):
# fmt: off
EXPECTED_RESULT = """ Folks, if you watch the show, you know, I spent a lot of time right over there. Patiently and astutely scrutinizing the boxwood and mahogany chest set of the day's biggest stories developing the central headline pawns, definitely maneuvering an oso topical night to F6, fainting a classic Sicilian, nade door variation on the news, all the while seeing eight moves deep and patiently marshalling the latest press releases into a fisher's shows in Lip Nitsky attack that culminates in the elegant lethal slow-played, all-passant checkmate that is my nightly monologue. But sometimes, sometimes, folks, I. CHEERING AND APPLAUSE Sometimes I startle away, cubside down in the monkey bars of a condemned playground on a super fun site. Get all hept up on goofballs. Rummage that were discarded tag bag of defective toys. Yank out a fist bowl of disembodied doll limbs, toss them on a stained kid's place mat from a defunct dennies. set up a table inside a rusty cargo container down by the Wharf and challenged toothless drifters to the godless bughouse blitz of tournament that is my segment. Meanwhile."""
EXPECTED_RESULT = " Folks, if you watch the show, you know, I spent a lot of time right over there. Patiently and astutely scrutinizing the boxwood and mahogany chest set of the day's biggest stories developing the central headline pawns, definitely maneuvering an oso topical night to F6, fainting a classic Sicilian, nade door variation on the news, all the while seeing eight moves deep and patiently marshalling the latest press releases into a fisher's shows in Lip Nitsky attack that culminates in the elegant lethal slow-played, all-passant checkmate that is my nightly monologue. But sometimes, sometimes, folks, I. CHEERING AND APPLAUSE Sometimes I startle away, cubside down in the monkey bars of a condemned playground on a super fun site. Get all hept up on goofballs. Rummage that were discarded tag bag of defective toys. Yank out a fist bowl of disembodied doll limbs, toss them on Saturday, Rusty Cargo, container down by the Wharf, and challenge toothless drifters to the godless bughouse lets of tournament that is my segment. MUSIC Meanwhile!"
# fmt: on
processor = AutoProcessor.from_pretrained("openai/whisper-tiny.en")

30
tests/pipelines/test_pipelines_document_question_answering.py
View File

@ -103,7 +103,9 @@ class DocumentQuestionAnsweringPipelineTests(unittest.TestCase):
@require_detectron2
@require_pytesseract
def test_small_model_pt(self):
dqa_pipeline = pipeline("document-question-answering", model="hf-internal-testing/tiny-random-layoutlmv2")
dqa_pipeline = pipeline(
"document-question-answering", model="hf-internal-testing/tiny-random-layoutlmv2-for-dqa-test"
)
image = INVOICE_URL
question = "How many cats are there?"
@ -253,19 +255,19 @@ class DocumentQuestionAnsweringPipelineTests(unittest.TestCase):
outputs = dqa_pipeline(image=image, question=question, top_k=2)
self.assertEqual(
nested_simplify(outputs, decimals=4),
nested_simplify(outputs, decimals=3),
[
{"score": 0.4251, "answer": "us-001", "start": 16, "end": 16},
{"score": 0.0819, "answer": "1110212019", "start": 23, "end": 23},
{"score": 0.425, "answer": "us-001", "start": 16, "end": 16},
{"score": 0.082, "answer": "1110212019", "start": 23, "end": 23},
],
)
outputs = dqa_pipeline({"image": image, "question": question}, top_k=2)
self.assertEqual(
nested_simplify(outputs, decimals=4),
nested_simplify(outputs, decimals=3),
[
{"score": 0.4251, "answer": "us-001", "start": 16, "end": 16},
{"score": 0.0819, "answer": "1110212019", "start": 23, "end": 23},
{"score": 0.425, "answer": "us-001", "start": 16, "end": 16},
{"score": 0.082, "answer": "1110212019", "start": 23, "end": 23},
],
)
@ -273,11 +275,11 @@ class DocumentQuestionAnsweringPipelineTests(unittest.TestCase):
[{"image": image, "question": question}, {"image": image, "question": question}], top_k=2
)
self.assertEqual(
nested_simplify(outputs, decimals=4),
nested_simplify(outputs, decimals=3),
[
[
{"score": 0.4251, "answer": "us-001", "start": 16, "end": 16},
{"score": 0.0819, "answer": "1110212019", "start": 23, "end": 23},
{"score": 0.425, "answer": "us-001", "start": 16, "end": 16},
{"score": 0.082, "answer": "1110212019", "start": 23, "end": 23},
]
]
* 2,
@ -288,10 +290,10 @@ class DocumentQuestionAnsweringPipelineTests(unittest.TestCase):
# This model should also work if `image` is set to None
outputs = dqa_pipeline({"image": None, "word_boxes": word_boxes, "question": question}, top_k=2)
self.assertEqual(
nested_simplify(outputs, decimals=4),
nested_simplify(outputs, decimals=3),
[
{"score": 0.4251, "answer": "us-001", "start": 16, "end": 16},
{"score": 0.0819, "answer": "1110212019", "start": 23, "end": 23},
{"score": 0.425, "answer": "us-001", "start": 16, "end": 16},
{"score": 0.082, "answer": "1110212019", "start": 23, "end": 23},
],
)
@ -355,7 +357,7 @@ class DocumentQuestionAnsweringPipelineTests(unittest.TestCase):
"document-question-answering",
model="naver-clova-ix/donut-base-finetuned-docvqa",
tokenizer=AutoTokenizer.from_pretrained("naver-clova-ix/donut-base-finetuned-docvqa"),
feature_extractor="naver-clova-ix/donut-base-finetuned-docvqa",
image_processor="naver-clova-ix/donut-base-finetuned-docvqa",
)
image = INVOICE_URL

12
tests/pipelines/test_pipelines_zero_shot_audio_classification.py
View File

@ -59,8 +59,8 @@ class ZeroShotAudioClassificationPipelineTests(unittest.TestCase):
self.assertEqual(
nested_simplify(output),
[
{"score": 0.999, "label": "Sound of a dog"},
{"score": 0.001, "label": "Sound of vaccum cleaner"},
{"score": 1.0, "label": "Sound of a dog"},
{"score": 0.0, "label": "Sound of vaccum cleaner"},
],
)
@ -69,8 +69,8 @@ class ZeroShotAudioClassificationPipelineTests(unittest.TestCase):
nested_simplify(output),
[
[
{"score": 0.999, "label": "Sound of a dog"},
{"score": 0.001, "label": "Sound of vaccum cleaner"},
{"score": 1.0, "label": "Sound of a dog"},
{"score": 0.0, "label": "Sound of vaccum cleaner"},
],
]
* 5,
@ -82,8 +82,8 @@ class ZeroShotAudioClassificationPipelineTests(unittest.TestCase):
nested_simplify(output),
[
[
{"score": 0.999, "label": "Sound of a dog"},
{"score": 0.001, "label": "Sound of vaccum cleaner"},
{"score": 1.0, "label": "Sound of a dog"},
{"score": 0.0, "label": "Sound of vaccum cleaner"},
],
]
* 5,