Skip to content

[Pipelines] Add revision tag to all default pipelines #17667

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
patrickvonplaten merged 10 commits into from
Jun 30, 2022
Merged

[Pipelines] Add revision tag to all default pipelines #17667

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
d54a4cb
trigger test failure
patrickvonplaten Jun 10, 2022
ca6a820
upload revision poc
patrickvonplaten Jun 10, 2022
3623fed
Update src/transformers/pipelines/base.py
patrickvonplaten Jun 21, 2022
38d686d
up
patrickvonplaten Jun 24, 2022
ab5324e
add test
patrickvonplaten Jun 30, 2022
618aee5
Merge branch 'main' of https://github.com/huggingface/transformers in…
patrickvonplaten Jun 30, 2022
2fcd9b6
Merge branch 'revision_tags_for_default_pipeline' of https://github.c…
patrickvonplaten Jun 30, 2022
24205b8
correct some stuff
patrickvonplaten Jun 30, 2022
4e4ff87
Update src/transformers/pipelines/__init__.py
patrickvonplaten Jun 30, 2022
f8ef027
correct require flag
patrickvonplaten Jun 30, 2022
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
79 changes: 44 additions & 35 deletions src/transformers/pipelines/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,7 +30,7 @@
from ..models.auto.tokenization_auto import TOKENIZER_MAPPING, AutoTokenizer
from ..tokenization_utils import PreTrainedTokenizer
from ..tokenization_utils_fast import PreTrainedTokenizerFast
from ..utils import http_get, is_tf_available, is_torch_available, logging
from ..utils import HUGGINGFACE_CO_RESOLVE_ENDPOINT, http_get, is_tf_available, is_torch_available, logging
from .audio_classification import AudioClassificationPipeline
from .automatic_speech_recognition import AutomaticSpeechRecognitionPipeline
from .base import (
Expand All @@ -41,7 +41,7 @@
Pipeline,
PipelineDataFormat,
PipelineException,
get_default_model,
get_default_model_and_revision,
infer_framework_load_model,
)
from .conversational import Conversation, ConversationalPipeline
Expand Down Expand Up @@ -131,21 +131,21 @@
"impl": AudioClassificationPipeline,
"tf": (),
"pt": (AutoModelForAudioClassification,) if is_torch_available() else (),
"default": {"model": {"pt": "superb/wav2vec2-base-superb-ks"}},
"default": {"model": {"pt": ("superb/wav2vec2-base-superb-ks", "372e048")}},
"type": "audio",
},
"automatic-speech-recognition": {
"impl": AutomaticSpeechRecognitionPipeline,
"tf": (),
"pt": (AutoModelForCTC, AutoModelForSpeechSeq2Seq) if is_torch_available() else (),
"default": {"model": {"pt": "facebook/wav2vec2-base-960h"}},
"default": {"model": {"pt": ("facebook/wav2vec2-base-960h", "55bb623")}},
"type": "multimodal",
},
"feature-extraction": {
"impl": FeatureExtractionPipeline,
"tf": (TFAutoModel,) if is_tf_available() else (),
"pt": (AutoModel,) if is_torch_available() else (),
"default": {"model": {"pt": "distilbert-base-cased", "tf": "distilbert-base-cased"}},
"default": {"model": {"pt": ("distilbert-base-cased", "935ac13"), "tf": ("distilbert-base-cased", "935ac13")}},
"type": "multimodal",
},
"text-classification": {
Expand All @@ -154,8 +154,8 @@
"pt": (AutoModelForSequenceClassification,) if is_torch_available() else (),
"default": {
"model": {
"pt": "distilbert-base-uncased-finetuned-sst-2-english",
"tf": "distilbert-base-uncased-finetuned-sst-2-english",
"pt": ("distilbert-base-uncased-finetuned-sst-2-english", "af0f99b"),
"tf": ("distilbert-base-uncased-finetuned-sst-2-english", "af0f99b"),
},
},
"type": "text",
Expand All @@ -166,8 +166,8 @@
"pt": (AutoModelForTokenClassification,) if is_torch_available() else (),
"default": {
"model": {
"pt": "dbmdz/bert-large-cased-finetuned-conll03-english",
"tf": "dbmdz/bert-large-cased-finetuned-conll03-english",
"pt": ("dbmdz/bert-large-cased-finetuned-conll03-english", "f2482bf"),
"tf": ("dbmdz/bert-large-cased-finetuned-conll03-english", "f2482bf"),
},
},
"type": "text",
Expand All @@ -177,7 +177,10 @@
"tf": (TFAutoModelForQuestionAnswering,) if is_tf_available() else (),
"pt": (AutoModelForQuestionAnswering,) if is_torch_available() else (),
"default": {
"model": {"pt": "distilbert-base-cased-distilled-squad", "tf": "distilbert-base-cased-distilled-squad"},
"model": {
"pt": ("distilbert-base-cased-distilled-squad", "626af31"),
"tf": ("distilbert-base-cased-distilled-squad", "626af31"),
},
},
"type": "text",
},
Expand All @@ -187,9 +190,8 @@
"tf": (TFAutoModelForTableQuestionAnswering,) if is_tf_available() else (),
"default": {
"model": {
"pt": "google/tapas-base-finetuned-wtq",
"tokenizer": "google/tapas-base-finetuned-wtq",
Copy link
Contributor Author

@patrickvonplaten patrickvonplaten Jun 30, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

deleted tokenizer because of https://github.com/huggingface/transformers/pull/17667/files#r910987282

"tf": "google/tapas-base-finetuned-wtq",
"pt": ("google/tapas-base-finetuned-wtq", "69ceee2"),
"tf": ("google/tapas-base-finetuned-wtq", "69ceee2"),
},
},
"type": "text",
Expand All @@ -199,26 +201,22 @@
"pt": (AutoModelForVisualQuestionAnswering,) if is_torch_available() else (),
"tf": (),
"default": {
"model": {
"pt": "dandelin/vilt-b32-finetuned-vqa",
"tokenizer": "dandelin/vilt-b32-finetuned-vqa",
"feature_extractor": "dandelin/vilt-b32-finetuned-vqa",
Comment on lines -204 to -205
Copy link
Contributor Author

@patrickvonplaten patrickvonplaten Jun 30, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

deleted tokenizer and feature extractor because of https://github.com/huggingface/transformers/pull/17667/files#r910987282

},
"model": {"pt": ("dandelin/vilt-b32-finetuned-vqa", "4355f59")},
},
"type": "multimodal",
},
"fill-mask": {
"impl": FillMaskPipeline,
"tf": (TFAutoModelForMaskedLM,) if is_tf_available() else (),
"pt": (AutoModelForMaskedLM,) if is_torch_available() else (),
"default": {"model": {"pt": "distilroberta-base", "tf": "distilroberta-base"}},
"default": {"model": {"pt": ("distilroberta-base", "ec58a5b"), "tf": ("distilroberta-base", "ec58a5b")}},
"type": "text",
},
"summarization": {
"impl": SummarizationPipeline,
"tf": (TFAutoModelForSeq2SeqLM,) if is_tf_available() else (),
"pt": (AutoModelForSeq2SeqLM,) if is_torch_available() else (),
"default": {"model": {"pt": "sshleifer/distilbart-cnn-12-6", "tf": "t5-small"}},
"default": {"model": {"pt": ("sshleifer/distilbart-cnn-12-6", "a4f8f3e"), "tf": ("t5-small", "d769bba")}},
"type": "text",
},
# This task is a special case as it's parametrized by SRC, TGT languages.
Expand All @@ -227,70 +225,76 @@
"tf": (TFAutoModelForSeq2SeqLM,) if is_tf_available() else (),
"pt": (AutoModelForSeq2SeqLM,) if is_torch_available() else (),
"default": {
("en", "fr"): {"model": {"pt": "t5-base", "tf": "t5-base"}},
("en", "de"): {"model": {"pt": "t5-base", "tf": "t5-base"}},
("en", "ro"): {"model": {"pt": "t5-base", "tf": "t5-base"}},
("en", "fr"): {"model": {"pt": ("t5-base", "686f1db"), "tf": ("t5-base", "686f1db")}},
("en", "de"): {"model": {"pt": ("t5-base", "686f1db"), "tf": ("t5-base", "686f1db")}},
("en", "ro"): {"model": {"pt": ("t5-base", "686f1db"), "tf": ("t5-base", "686f1db")}},
},
"type": "text",
},
"text2text-generation": {
"impl": Text2TextGenerationPipeline,
"tf": (TFAutoModelForSeq2SeqLM,) if is_tf_available() else (),
"pt": (AutoModelForSeq2SeqLM,) if is_torch_available() else (),
"default": {"model": {"pt": "t5-base", "tf": "t5-base"}},
"default": {"model": {"pt": ("t5-base", "686f1db"), "tf": ("t5-base", "686f1db")}},
"type": "text",
},
"text-generation": {
"impl": TextGenerationPipeline,
"tf": (TFAutoModelForCausalLM,) if is_tf_available() else (),
"pt": (AutoModelForCausalLM,) if is_torch_available() else (),
"default": {"model": {"pt": "gpt2", "tf": "gpt2"}},
"default": {"model": {"pt": ("gpt2", "6c0e608"), "tf": ("gpt2", "6c0e608")}},
"type": "text",
},
"zero-shot-classification": {
"impl": ZeroShotClassificationPipeline,
"tf": (TFAutoModelForSequenceClassification,) if is_tf_available() else (),
"pt": (AutoModelForSequenceClassification,) if is_torch_available() else (),
"default": {
"model": {"pt": "facebook/bart-large-mnli", "tf": "roberta-large-mnli"},
"config": {"pt": "facebook/bart-large-mnli", "tf": "roberta-large-mnli"},
"tokenizer": {"pt": "facebook/bart-large-mnli", "tf": "roberta-large-mnli"},
Copy link
Contributor Author

@patrickvonplaten patrickvonplaten Jun 30, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@Narsil before merging this PR, I'd love to have your opinion here. I've checked a the pipeline function and I don't think a "default" tokenizer is never used. If no repo_id is provided it seems like for tokenizer or feature extractor it's always the model id that is used and never the tokenizer id => to me it looks like this is dead code here. Can you confirm?

"model": {"pt": ("facebook/bart-large-mnli", "c626438"), "tf": ("roberta-large-mnli", "130fb28")},
"config": {"pt": ("facebook/bart-large-mnli", "c626438"), "tf": ("roberta-large-mnli", "130fb28")},
},
"type": "text",
},
"zero-shot-image-classification": {
"impl": ZeroShotImageClassificationPipeline,
"tf": (TFAutoModel,) if is_tf_available() else (),
"pt": (AutoModel,) if is_torch_available() else (),
"default": {"model": {"pt": "openai/clip-vit-base-patch32", "tf": "openai/clip-vit-base-patch32"}},
"default": {
"model": {
"pt": ("openai/clip-vit-base-patch32", "f4881ba"),
"tf": ("openai/clip-vit-base-patch32", "f4881ba"),
}
},
"type": "multimodal",
},
"conversational": {
"impl": ConversationalPipeline,
"tf": (TFAutoModelForSeq2SeqLM, TFAutoModelForCausalLM) if is_tf_available() else (),
"pt": (AutoModelForSeq2SeqLM, AutoModelForCausalLM) if is_torch_available() else (),
"default": {"model": {"pt": "microsoft/DialoGPT-medium", "tf": "microsoft/DialoGPT-medium"}},
"default": {
"model": {"pt": ("microsoft/DialoGPT-medium", "8bada3b"), "tf": ("microsoft/DialoGPT-medium", "8bada3b")}
},
"type": "text",
},
"image-classification": {
"impl": ImageClassificationPipeline,
"tf": (),
"pt": (AutoModelForImageClassification,) if is_torch_available() else (),
"default": {"model": {"pt": "google/vit-base-patch16-224"}},
"default": {"model": {"pt": ("google/vit-base-patch16-224", "5dca96d")}},
"type": "image",
},
"image-segmentation": {
"impl": ImageSegmentationPipeline,
"tf": (),
"pt": (AutoModelForImageSegmentation, AutoModelForSemanticSegmentation) if is_torch_available() else (),
"default": {"model": {"pt": "facebook/detr-resnet-50-panoptic"}},
"default": {"model": {"pt": ("facebook/detr-resnet-50-panoptic", "fc15262")}},
"type": "image",
},
"object-detection": {
"impl": ObjectDetectionPipeline,
"tf": (),
"pt": (AutoModelForObjectDetection,) if is_torch_available() else (),
"default": {"model": {"pt": "facebook/detr-resnet-50"}},
"default": {"model": {"pt": ("facebook/detr-resnet-50", "2729413")}},
"type": "image",
},
}
Expand Down Expand Up @@ -545,8 +549,13 @@ def pipeline(
# Use default model/config/tokenizer for the task if no model is provided
if model is None:
# At that point framework might still be undetermined
model = get_default_model(targeted_task, framework, task_options)
logger.warning(f"No model was supplied, defaulted to {model} (https://huggingface.co/{model})")
model, default_revision = get_default_model_and_revision(targeted_task, framework, task_options)
revision = revision if revision is not None else default_revision
logger.warning(
f"No model was supplied, defaulted to {model} and revision"
f" {revision} ({HUGGINGFACE_CO_RESOLVE_ENDPOINT}/{model}).\n"
"Using a pipeline without specifying a model name and revision in production is not recommended."
)

# Retrieve use_auth_token and add it to model_kwargs to be used in .from_pretrained
model_kwargs["use_auth_token"] = model_kwargs.get("use_auth_token", use_auth_token)
Expand Down
4 changes: 3 additions & 1 deletion src/transformers/pipelines/base.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -341,7 +341,9 @@ def get_framework(model, revision: Optional[str] = None):
return framework


def get_default_model(targeted_task: Dict, framework: Optional[str], task_options: Optional[Any]) -> str:
def get_default_model_and_revision(
targeted_task: Dict, framework: Optional[str], task_options: Optional[Any]
) -> Union[str, Tuple[str, str]]:
"""
Select a default model to use for a given task. Defaults to pytorch if ambiguous.

Expand Down
Loading