huggingface-hub 0.21.2__py3-none-any.whl → 0.22.0__py3-none-any.whl

huggingface_hub/__init__.py

@@ -46,7 +46,7 @@ import sys
 from typing import TYPE_CHECKING
 
 
-__version__ = "0.21.2"
+__version__ = "0.22.0"
 
 # Alphabetical order of definitions is ensured in tests
 # WARNING: any comment added in this dictionary definition will be lost when
@@ -259,6 +259,114 @@ _SUBMOD_ATTRS = {
     "inference._generated._async_client": [
         "AsyncInferenceClient",
     ],
+    "inference._generated.types": [
+        "AudioClassificationInput",
+        "AudioClassificationOutputElement",
+        "AudioClassificationParameters",
+        "AudioToAudioInput",
+        "AudioToAudioOutputElement",
+        "AutomaticSpeechRecognitionGenerationParameters",
+        "AutomaticSpeechRecognitionInput",
+        "AutomaticSpeechRecognitionOutput",
+        "AutomaticSpeechRecognitionOutputChunk",
+        "AutomaticSpeechRecognitionParameters",
+        "ChatCompletionInput",
+        "ChatCompletionInputMessage",
+        "ChatCompletionOutput",
+        "ChatCompletionOutputChoice",
+        "ChatCompletionOutputChoiceMessage",
+        "ChatCompletionStreamOutput",
+        "ChatCompletionStreamOutputChoice",
+        "ChatCompletionStreamOutputDelta",
+        "DepthEstimationInput",
+        "DepthEstimationOutput",
+        "DocumentQuestionAnsweringInput",
+        "DocumentQuestionAnsweringInputData",
+        "DocumentQuestionAnsweringOutputElement",
+        "DocumentQuestionAnsweringParameters",
+        "FeatureExtractionInput",
+        "FillMaskInput",
+        "FillMaskOutputElement",
+        "FillMaskParameters",
+        "ImageClassificationInput",
+        "ImageClassificationOutputElement",
+        "ImageClassificationParameters",
+        "ImageSegmentationInput",
+        "ImageSegmentationOutputElement",
+        "ImageSegmentationParameters",
+        "ImageToImageInput",
+        "ImageToImageOutput",
+        "ImageToImageParameters",
+        "ImageToImageTargetSize",
+        "ImageToTextGenerationParameters",
+        "ImageToTextInput",
+        "ImageToTextOutput",
+        "ImageToTextParameters",
+        "ObjectDetectionBoundingBox",
+        "ObjectDetectionInput",
+        "ObjectDetectionOutputElement",
+        "ObjectDetectionParameters",
+        "QuestionAnsweringInput",
+        "QuestionAnsweringInputData",
+        "QuestionAnsweringOutputElement",
+        "QuestionAnsweringParameters",
+        "SentenceSimilarityInput",
+        "SentenceSimilarityInputData",
+        "SummarizationGenerationParameters",
+        "SummarizationInput",
+        "SummarizationOutput",
+        "TableQuestionAnsweringInput",
+        "TableQuestionAnsweringInputData",
+        "TableQuestionAnsweringOutputElement",
+        "Text2TextGenerationInput",
+        "Text2TextGenerationOutput",
+        "Text2TextGenerationParameters",
+        "TextClassificationInput",
+        "TextClassificationOutputElement",
+        "TextClassificationParameters",
+        "TextGenerationInput",
+        "TextGenerationOutput",
+        "TextGenerationOutputDetails",
+        "TextGenerationOutputSequenceDetails",
+        "TextGenerationOutputToken",
+        "TextGenerationParameters",
+        "TextGenerationPrefillToken",
+        "TextGenerationStreamDetails",
+        "TextGenerationStreamOutput",
+        "TextToAudioGenerationParameters",
+        "TextToAudioInput",
+        "TextToAudioOutput",
+        "TextToAudioParameters",
+        "TextToImageInput",
+        "TextToImageOutput",
+        "TextToImageParameters",
+        "TextToImageTargetSize",
+        "TokenClassificationInput",
+        "TokenClassificationOutputElement",
+        "TokenClassificationParameters",
+        "TranslationGenerationParameters",
+        "TranslationInput",
+        "TranslationOutput",
+        "VideoClassificationInput",
+        "VideoClassificationOutputElement",
+        "VideoClassificationParameters",
+        "VisualQuestionAnsweringInput",
+        "VisualQuestionAnsweringInputData",
+        "VisualQuestionAnsweringOutputElement",
+        "VisualQuestionAnsweringParameters",
+        "ZeroShotClassificationInput",
+        "ZeroShotClassificationInputData",
+        "ZeroShotClassificationOutputElement",
+        "ZeroShotClassificationParameters",
+        "ZeroShotImageClassificationInput",
+        "ZeroShotImageClassificationInputData",
+        "ZeroShotImageClassificationOutputElement",
+        "ZeroShotImageClassificationParameters",
+        "ZeroShotObjectDetectionBoundingBox",
+        "ZeroShotObjectDetectionInput",
+        "ZeroShotObjectDetectionInputData",
+        "ZeroShotObjectDetectionOutputElement",
+    ],
     "inference_api": [
         "InferenceApi",
     ],
@@ -613,6 +721,114 @@ if TYPE_CHECKING:  # pragma: no cover
         InferenceTimeoutError,  # noqa: F401
     )
     from .inference._generated._async_client import AsyncInferenceClient  # noqa: F401
+    from .inference._generated.types import (
+        AudioClassificationInput,  # noqa: F401
+        AudioClassificationOutputElement,  # noqa: F401
+        AudioClassificationParameters,  # noqa: F401
+        AudioToAudioInput,  # noqa: F401
+        AudioToAudioOutputElement,  # noqa: F401
+        AutomaticSpeechRecognitionGenerationParameters,  # noqa: F401
+        AutomaticSpeechRecognitionInput,  # noqa: F401
+        AutomaticSpeechRecognitionOutput,  # noqa: F401
+        AutomaticSpeechRecognitionOutputChunk,  # noqa: F401
+        AutomaticSpeechRecognitionParameters,  # noqa: F401
+        ChatCompletionInput,  # noqa: F401
+        ChatCompletionInputMessage,  # noqa: F401
+        ChatCompletionOutput,  # noqa: F401
+        ChatCompletionOutputChoice,  # noqa: F401
+        ChatCompletionOutputChoiceMessage,  # noqa: F401
+        ChatCompletionStreamOutput,  # noqa: F401
+        ChatCompletionStreamOutputChoice,  # noqa: F401
+        ChatCompletionStreamOutputDelta,  # noqa: F401
+        DepthEstimationInput,  # noqa: F401
+        DepthEstimationOutput,  # noqa: F401
+        DocumentQuestionAnsweringInput,  # noqa: F401
+        DocumentQuestionAnsweringInputData,  # noqa: F401
+        DocumentQuestionAnsweringOutputElement,  # noqa: F401
+        DocumentQuestionAnsweringParameters,  # noqa: F401
+        FeatureExtractionInput,  # noqa: F401
+        FillMaskInput,  # noqa: F401
+        FillMaskOutputElement,  # noqa: F401
+        FillMaskParameters,  # noqa: F401
+        ImageClassificationInput,  # noqa: F401
+        ImageClassificationOutputElement,  # noqa: F401
+        ImageClassificationParameters,  # noqa: F401
+        ImageSegmentationInput,  # noqa: F401
+        ImageSegmentationOutputElement,  # noqa: F401
+        ImageSegmentationParameters,  # noqa: F401
+        ImageToImageInput,  # noqa: F401
+        ImageToImageOutput,  # noqa: F401
+        ImageToImageParameters,  # noqa: F401
+        ImageToImageTargetSize,  # noqa: F401
+        ImageToTextGenerationParameters,  # noqa: F401
+        ImageToTextInput,  # noqa: F401
+        ImageToTextOutput,  # noqa: F401
+        ImageToTextParameters,  # noqa: F401
+        ObjectDetectionBoundingBox,  # noqa: F401
+        ObjectDetectionInput,  # noqa: F401
+        ObjectDetectionOutputElement,  # noqa: F401
+        ObjectDetectionParameters,  # noqa: F401
+        QuestionAnsweringInput,  # noqa: F401
+        QuestionAnsweringInputData,  # noqa: F401
+        QuestionAnsweringOutputElement,  # noqa: F401
+        QuestionAnsweringParameters,  # noqa: F401
+        SentenceSimilarityInput,  # noqa: F401
+        SentenceSimilarityInputData,  # noqa: F401
+        SummarizationGenerationParameters,  # noqa: F401
+        SummarizationInput,  # noqa: F401
+        SummarizationOutput,  # noqa: F401
+        TableQuestionAnsweringInput,  # noqa: F401
+        TableQuestionAnsweringInputData,  # noqa: F401
+        TableQuestionAnsweringOutputElement,  # noqa: F401
+        Text2TextGenerationInput,  # noqa: F401
+        Text2TextGenerationOutput,  # noqa: F401
+        Text2TextGenerationParameters,  # noqa: F401
+        TextClassificationInput,  # noqa: F401
+        TextClassificationOutputElement,  # noqa: F401
+        TextClassificationParameters,  # noqa: F401
+        TextGenerationInput,  # noqa: F401
+        TextGenerationOutput,  # noqa: F401
+        TextGenerationOutputDetails,  # noqa: F401
+        TextGenerationOutputSequenceDetails,  # noqa: F401
+        TextGenerationOutputToken,  # noqa: F401
+        TextGenerationParameters,  # noqa: F401
+        TextGenerationPrefillToken,  # noqa: F401
+        TextGenerationStreamDetails,  # noqa: F401
+        TextGenerationStreamOutput,  # noqa: F401
+        TextToAudioGenerationParameters,  # noqa: F401
+        TextToAudioInput,  # noqa: F401
+        TextToAudioOutput,  # noqa: F401
+        TextToAudioParameters,  # noqa: F401
+        TextToImageInput,  # noqa: F401
+        TextToImageOutput,  # noqa: F401
+        TextToImageParameters,  # noqa: F401
+        TextToImageTargetSize,  # noqa: F401
+        TokenClassificationInput,  # noqa: F401
+        TokenClassificationOutputElement,  # noqa: F401
+        TokenClassificationParameters,  # noqa: F401
+        TranslationGenerationParameters,  # noqa: F401
+        TranslationInput,  # noqa: F401
+        TranslationOutput,  # noqa: F401
+        VideoClassificationInput,  # noqa: F401
+        VideoClassificationOutputElement,  # noqa: F401
+        VideoClassificationParameters,  # noqa: F401
+        VisualQuestionAnsweringInput,  # noqa: F401
+        VisualQuestionAnsweringInputData,  # noqa: F401
+        VisualQuestionAnsweringOutputElement,  # noqa: F401
+        VisualQuestionAnsweringParameters,  # noqa: F401
+        ZeroShotClassificationInput,  # noqa: F401
+        ZeroShotClassificationInputData,  # noqa: F401
+        ZeroShotClassificationOutputElement,  # noqa: F401
+        ZeroShotClassificationParameters,  # noqa: F401
+        ZeroShotImageClassificationInput,  # noqa: F401
+        ZeroShotImageClassificationInputData,  # noqa: F401
+        ZeroShotImageClassificationOutputElement,  # noqa: F401
+        ZeroShotImageClassificationParameters,  # noqa: F401
+        ZeroShotObjectDetectionBoundingBox,  # noqa: F401
+        ZeroShotObjectDetectionInput,  # noqa: F401
+        ZeroShotObjectDetectionInputData,  # noqa: F401
+        ZeroShotObjectDetectionOutputElement,  # noqa: F401
+    )
     from .inference_api import InferenceApi  # noqa: F401
     from .keras_mixin import (
         KerasModelHubMixin,  # noqa: F401

huggingface_hub/_commit_api.py

@@ -1,6 +1,7 @@
 """
 Type definitions and utilities for the `create_commit` API
 """
+
 import base64
 import io
 import os
@@ -21,7 +22,6 @@ from .file_download import hf_hub_url
 from .lfs import UploadInfo, lfs_upload, post_lfs_batch_info
 from .utils import (
     EntryNotFoundError,
-    build_hf_headers,
     chunk_iterable,
     hf_raise_for_status,
     logging,
@@ -318,7 +318,7 @@ def _upload_lfs_files(
     additions: List[CommitOperationAdd],
     repo_type: str,
     repo_id: str,
-    token: Optional[str],
+    headers: Dict[str, str],
     endpoint: Optional[str] = None,
     num_threads: int = 5,
     revision: Optional[str] = None,
@@ -337,8 +337,8 @@ def _upload_lfs_files(
         repo_id (`str`):
             A namespace (user or an organization) and a repo name separated
             by a `/`.
-        token (`str`, *optional*):
-            An authentication token ( See https://huggingface.co/settings/tokens )
+        headers (`Dict[str, str]`):
+            Headers to use for the request, including authorization headers and user agent.
         num_threads (`int`, *optional*):
             The number of concurrent threads to use when uploading. Defaults to 5.
         revision (`str`, *optional*):
@@ -359,11 +359,12 @@ def _upload_lfs_files(
     for chunk in chunk_iterable(additions, chunk_size=256):
         batch_actions_chunk, batch_errors_chunk = post_lfs_batch_info(
             upload_infos=[op.upload_info for op in chunk],
-            token=token,
             repo_id=repo_id,
             repo_type=repo_type,
             revision=revision,
             endpoint=endpoint,
+            headers=headers,
+            token=None,  # already passed in 'headers'
         )
 
         # If at least 1 error, we do not retrieve information for other chunks
@@ -398,7 +399,7 @@ def _upload_lfs_files(
     def _wrapped_lfs_upload(batch_action) -> None:
         try:
             operation = oid2addop[batch_action["oid"]]
-            lfs_upload(operation=operation, lfs_batch_action=batch_action, token=token)
+            lfs_upload(operation=operation, lfs_batch_action=batch_action, headers=headers)
         except Exception as exc:
             raise RuntimeError(f"Error while uploading '{operation.path_in_repo}' to the Hub.") from exc
 
@@ -442,7 +443,7 @@ def _fetch_upload_modes(
     additions: Iterable[CommitOperationAdd],
     repo_type: str,
     repo_id: str,
-    token: Optional[str],
+    headers: Dict[str, str],
     revision: str,
     endpoint: Optional[str] = None,
     create_pr: bool = False,
@@ -461,8 +462,8 @@ def _fetch_upload_modes(
         repo_id (`str`):
             A namespace (user or an organization) and a repo name separated
             by a `/`.
-        token (`str`, *optional*):
-            An authentication token ( See https://huggingface.co/settings/tokens )
+        headers (`Dict[str, str]`):
+            Headers to use for the request, including authorization headers and user agent.
         revision (`str`):
             The git revision to upload the files to. Can be any valid git revision.
         gitignore_content (`str`, *optional*):
@@ -477,7 +478,6 @@ def _fetch_upload_modes(
             If the Hub API response is improperly formatted.
     """
     endpoint = endpoint if endpoint is not None else ENDPOINT
-    headers = build_hf_headers(token=token)
 
     # Fetch upload mode (LFS or regular) chunk by chunk.
     upload_modes: Dict[str, UploadMode] = {}
@@ -526,7 +526,7 @@ def _fetch_files_to_copy(
     copies: Iterable[CommitOperationCopy],
     repo_type: str,
     repo_id: str,
-    token: Optional[str],
+    headers: Dict[str, str],
     revision: str,
     endpoint: Optional[str] = None,
 ) -> Dict[Tuple[str, Optional[str]], Union["RepoFile", bytes]]:
@@ -545,8 +545,8 @@ def _fetch_files_to_copy(
         repo_id (`str`):
             A namespace (user or an organization) and a repo name separated
             by a `/`.
-        token (`str`, *optional*):
-            An authentication token ( See https://huggingface.co/settings/tokens )
+        headers (`Dict[str, str]`):
+            Headers to use for the request, including authorization headers and user agent.
         revision (`str`):
             The git revision to upload the files to. Can be any valid git revision.
 
@@ -562,7 +562,7 @@ def _fetch_files_to_copy(
     """
     from .hf_api import HfApi, RepoFolder
 
-    hf_api = HfApi(endpoint=endpoint, token=token)
+    hf_api = HfApi(endpoint=endpoint, headers=headers)
     files_to_copy: Dict[Tuple[str, Optional[str]], Union["RepoFile", bytes]] = {}
     for src_revision, operations in groupby(copies, key=lambda op: op.src_revision):
         operations = list(operations)  # type: ignore
@@ -581,7 +581,6 @@ def _fetch_files_to_copy(
                     files_to_copy[(src_repo_file.path, src_revision)] = src_repo_file
                 else:
                     # TODO: (optimization) download regular files to copy concurrently
-                    headers = build_hf_headers(token=token)
                     url = hf_hub_url(
                         endpoint=endpoint,
                         repo_type=repo_type,

huggingface_hub/_inference_endpoints.py

@@ -2,7 +2,7 @@ import time
 from dataclasses import dataclass, field
 from datetime import datetime
 from enum import Enum
-from typing import TYPE_CHECKING, Dict, Optional
+from typing import TYPE_CHECKING, Dict, Optional, Union
 
 from .inference._client import InferenceClient
 from .inference._generated._async_client import AsyncInferenceClient
@@ -71,8 +71,9 @@ class InferenceEndpoint:
             The type of the Inference Endpoint (public, protected, private).
         raw (`Dict`):
             The raw dictionary data returned from the API.
-        token (`str`, *optional*):
-            Authentication token for the Inference Endpoint, if set when requesting the API.
+        token (`str` or `bool`, *optional*):
+            Authentication token for the Inference Endpoint, if set when requesting the API. Will default to the
+            locally saved token if not provided. Pass `token=False` if you don't want to send your token to the server.
 
     Example:
         ```python
@@ -120,12 +121,12 @@ class InferenceEndpoint:
     raw: Dict = field(repr=False)
 
     # Internal fields
-    _token: Optional[str] = field(repr=False, compare=False)
+    _token: Union[str, bool, None] = field(repr=False, compare=False)
     _api: "HfApi" = field(repr=False, compare=False)
 
     @classmethod
     def from_raw(
-        cls, raw: Dict, namespace: str, token: Optional[str] = None, api: Optional["HfApi"] = None
+        cls, raw: Dict, namespace: str, token: Union[str, bool, None] = None, api: Optional["HfApi"] = None
     ) -> "InferenceEndpoint":
         """Initialize object from raw dictionary."""
         if api is None:
@@ -230,7 +231,7 @@ class InferenceEndpoint:
         Returns:
             [`InferenceEndpoint`]: the same Inference Endpoint, mutated in place with the latest data.
         """
-        obj = self._api.get_inference_endpoint(name=self.name, namespace=self.namespace, token=self._token)
+        obj = self._api.get_inference_endpoint(name=self.name, namespace=self.namespace, token=self._token)  # type: ignore [arg-type]
         self.raw = obj.raw
         self._populate_from_raw()
         return self
@@ -295,7 +296,7 @@ class InferenceEndpoint:
             framework=framework,
             revision=revision,
             task=task,
-            token=self._token,
+            token=self._token,  # type: ignore [arg-type]
         )
 
         # Mutate current object
@@ -316,7 +317,7 @@ class InferenceEndpoint:
         Returns:
             [`InferenceEndpoint`]: the same Inference Endpoint, mutated in place with the latest data.
         """
-        obj = self._api.pause_inference_endpoint(name=self.name, namespace=self.namespace, token=self._token)
+        obj = self._api.pause_inference_endpoint(name=self.name, namespace=self.namespace, token=self._token)  # type: ignore [arg-type]
         self.raw = obj.raw
         self._populate_from_raw()
         return self
@@ -330,7 +331,7 @@ class InferenceEndpoint:
         Returns:
             [`InferenceEndpoint`]: the same Inference Endpoint, mutated in place with the latest data.
         """
-        obj = self._api.resume_inference_endpoint(name=self.name, namespace=self.namespace, token=self._token)
+        obj = self._api.resume_inference_endpoint(name=self.name, namespace=self.namespace, token=self._token)  # type: ignore [arg-type]
         self.raw = obj.raw
         self._populate_from_raw()
         return self
@@ -348,7 +349,7 @@ class InferenceEndpoint:
         Returns:
             [`InferenceEndpoint`]: the same Inference Endpoint, mutated in place with the latest data.
         """
-        obj = self._api.scale_to_zero_inference_endpoint(name=self.name, namespace=self.namespace, token=self._token)
+        obj = self._api.scale_to_zero_inference_endpoint(name=self.name, namespace=self.namespace, token=self._token)  # type: ignore [arg-type]
         self.raw = obj.raw
         self._populate_from_raw()
         return self
@@ -361,7 +362,7 @@ class InferenceEndpoint:
 
         This is an alias for [`HfApi.delete_inference_endpoint`].
         """
-        self._api.delete_inference_endpoint(name=self.name, namespace=self.namespace, token=self._token)
+        self._api.delete_inference_endpoint(name=self.name, namespace=self.namespace, token=self._token)  # type: ignore [arg-type]
 
     def _populate_from_raw(self) -> None:
         """Populate fields from raw dictionary.

huggingface_hub/_login.py

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """Contains methods to login to the Hub."""
+
 import os
 import subprocess
 from functools import partial

huggingface_hub/_multi_commits.py

@@ -13,6 +13,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """Contains utilities to multi-commits (i.e. push changes iteratively on a PR)."""
+
 import re
 from dataclasses import dataclass, field
 from typing import TYPE_CHECKING, Iterable, List, Optional, Set, Tuple, Union

huggingface_hub/_snapshot_download.py

@@ -53,6 +53,7 @@ def snapshot_download(
     ignore_patterns: Optional[Union[List[str], str]] = None,
     max_workers: int = 8,
     tqdm_class: Optional[base_tqdm] = None,
+    headers: Optional[Dict[str, str]] = None,
     endpoint: Optional[str] = None,
 ) -> str:
     """Download repo files.
@@ -120,6 +121,8 @@ def snapshot_download(
                 - If `True`, the token is read from the HuggingFace config
                   folder.
                 - If a string, it's used as the authentication token.
+        headers (`dict`, *optional*):
+            Additional headers to include in the request. Those headers take precedence over the others.
         local_files_only (`bool`, *optional*, defaults to `False`):
             If `True`, avoid downloading the file and return the path to the
             local cached file if it exists.
@@ -174,7 +177,11 @@ def snapshot_download(
         try:
             # if we have internet connection we want to list files to download
             api = HfApi(
-                library_name=library_name, library_version=library_version, user_agent=user_agent, endpoint=endpoint
+                library_name=library_name,
+                library_version=library_version,
+                user_agent=user_agent,
+                endpoint=endpoint,
+                headers=headers,
             )
             repo_info = api.repo_info(repo_id=repo_id, repo_type=repo_type, revision=revision, token=token)
         except (requests.exceptions.SSLError, requests.exceptions.ProxyError):
@@ -297,6 +304,7 @@ def snapshot_download(
             resume_download=resume_download,
             force_download=force_download,
             token=token,
+            headers=headers,
         )
 
     if HF_HUB_ENABLE_HF_TRANSFER:

huggingface_hub/_tensorboard_logger.py

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """Contains a logger to push training logs to the Hub, using Tensorboard."""
+
 from pathlib import Path
 from typing import TYPE_CHECKING, List, Optional, Union
 

huggingface_hub/_webhooks_payload.py

@@ -13,6 +13,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """Contains data structures to parse the webhooks payload."""
+
 from typing import List, Literal, Optional
 
 from pydantic import BaseModel

huggingface_hub/_webhooks_server.py

@@ -13,6 +13,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """Contains `WebhooksServer` and `webhook_endpoint` to create a webhook server easily."""
+
 import atexit
 import inspect
 import os

huggingface_hub/commands/_cli_utils.py

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 """Contains a utility for good-looking prints."""
+
 import os
 from typing import List, Union
 

huggingface_hub/commands/delete_cache.py

@@ -55,6 +55,7 @@ TODO: add "--limit" arg to limit to X repos ?
 TODO: add "-y" arg for immediate deletion ?
 See discussions in https://github.com/huggingface/huggingface_hub/issues/1025.
 """
+
 import os
 from argparse import Namespace, _SubParsersAction
 from functools import wraps

huggingface_hub/commands/download.py

@@ -35,6 +35,7 @@ Usage:
     # Download to local dir
     huggingface-cli download gpt2 --local-dir=./models/gpt2
 """
+
 import warnings
 from argparse import Namespace, _SubParsersAction
 from typing import List, Literal, Optional, Union

huggingface_hub/commands/env.py

@@ -16,6 +16,7 @@
 Usage:
     huggingface-cli env
 """
+
 from argparse import _SubParsersAction
 
 from ..utils import dump_environment_info

huggingface_hub/commands/scan_cache.py

@@ -20,6 +20,7 @@ Usage:
     huggingface-cli scan-cache -vvv
     huggingface-cli scan-cache --dir ~/.cache/huggingface/hub
 """
+
 import time
 from argparse import Namespace, _SubParsersAction
 from typing import Optional

huggingface_hub/commands/upload.py

@@ -42,6 +42,7 @@ Usage:
     # Schedule commits every 30 minutes
     huggingface-cli upload Wauplin/my-cool-model --every=30
 """
+
 import os
 import time
 import warnings

huggingface_hub/community.py

@@ -4,6 +4,7 @@ Data structures to interact with Discussions and Pull Requests on the Hub.
 See [the Discussions and Pull Requests guide](https://huggingface.co/docs/hub/repositories-pull-requests-discussions)
 for more information on Pull Requests, Discussions, and the community tab.
 """
+
 from dataclasses import dataclass
 from datetime import datetime
 from typing import List, Literal, Optional, Union

huggingface_hub/constants.py

@@ -52,7 +52,9 @@ HUGGINGFACE_CO_URL_HOME = "https://huggingface.co/"
 
 _staging_mode = _is_true(os.environ.get("HUGGINGFACE_CO_STAGING"))
 
-ENDPOINT = os.getenv("HF_ENDPOINT") or ("https://hub-ci.huggingface.co" if _staging_mode else "https://huggingface.co")
+_HF_DEFAULT_ENDPOINT = "https://huggingface.co"
+_HF_DEFAULT_STAGING_ENDPOINT = "https://hub-ci.huggingface.co"
+ENDPOINT = os.getenv("HF_ENDPOINT") or (_HF_DEFAULT_STAGING_ENDPOINT if _staging_mode else _HF_DEFAULT_ENDPOINT)
 
 HUGGINGFACE_CO_URL_TEMPLATE = ENDPOINT + "/{repo_id}/resolve/{revision}/{filename}"
 HUGGINGFACE_HEADER_X_REPO_COMMIT = "X-Repo-Commit"

huggingface_hub/errors.py

@@ -0,0 +1,38 @@
+"""Contains all custom errors."""
+
+from requests import HTTPError
+
+
+# INFERENCE CLIENT ERRORS
+
+
+class InferenceTimeoutError(HTTPError, TimeoutError):
+    """Error raised when a model is unavailable or the request times out."""
+
+
+# TEXT GENERATION ERRORS
+
+
+class TextGenerationError(HTTPError):
+    """Generic error raised if text-generation went wrong."""
+
+
+# Text Generation Inference Errors
+class ValidationError(TextGenerationError):
+    """Server-side validation error."""
+
+
+class GenerationError(TextGenerationError):
+    pass
+
+
+class OverloadedError(TextGenerationError):
+    pass
+
+
+class IncompleteGenerationError(TextGenerationError):
+    pass
+
+
+class UnknownError(TextGenerationError):
+    pass