pipeline support for device="mps" (or any other string)

Author: julien-c

src/transformers/pipelines/__init__.py

@@ -420,6 +420,7 @@ def pipeline(
     revision: Optional[str] = None,
     use_fast: bool = True,
     use_auth_token: Optional[Union[str, bool]] = None,
+    device: Optional[Union[int, str, "torch.device"]] = None,
     device_map=None,
     torch_dtype=None,
     trust_remote_code: Optional[bool] = None,
@@ -506,6 +507,9 @@ def pipeline(
         use_auth_token (`str` or *bool*, *optional*):
             The token to use as HTTP bearer authorization for remote files. If `True`, will use the token generated
             when running `transformers-cli login` (stored in `~/.huggingface`).
+        device (`int` or `str` or `torch.device`):
+            Sent directly as `model_kwargs` (just a simpler shortcut). Defines the device (*e.g.*, `"cpu"`, `"cuda:1"`,
+            `"mps"`, or a GPU ordinal rank like `1`) on which this pipeline will be allocated.
         device_map (`str` or `Dict[str, Union[int, str, torch.device]`, *optional*):
             Sent directly as `model_kwargs` (just a simpler shortcut). When `accelerate` library is present, set
             `device_map="auto"` to compute the most optimized `device_map` automatically. [More
@@ -800,4 +804,4 @@ def pipeline(
     if feature_extractor is not None:
         kwargs["feature_extractor"] = feature_extractor
 
-    return pipeline_class(model=model, framework=framework, task=task, **kwargs)
+    return pipeline_class(model=model, framework=framework, task=task, device=device, **kwargs)

src/transformers/pipelines/base.py

@@ -704,7 +704,7 @@ def predict(self, X):
             Reference to the object in charge of parsing supplied pipeline parameters.
         device (`int`, *optional*, defaults to -1):
             Device ordinal for CPU/GPU supports. Setting this to -1 will leverage CPU, a positive will run the model on
-            the associated CUDA device id. You can pass native `torch.device` too.
+            the associated CUDA device id. You can pass native `torch.device` or a `str` too.
         binary_output (`bool`, *optional*, defaults to `False`):
             Flag indicating if the output the pipeline should happen in a binary format (i.e., pickle) or as raw text.
 """
@@ -747,7 +747,7 @@ def __init__(
         framework: Optional[str] = None,
         task: str = "",
         args_parser: ArgumentHandler = None,
-        device: int = -1,
+        device: Union[int, str, "torch.device"] = -1,
         binary_output: bool = False,
         **kwargs,
     ):
@@ -763,11 +763,15 @@ def __init__(
         if is_torch_available() and isinstance(device, torch.device):
             self.device = device
         else:
-            self.device = device if framework == "tf" else torch.device("cpu" if device < 0 else f"cuda:{device}")
+            self.device = (
+                device
+                if framework == "tf"
+                else torch.device(device if type(device) == str else "cpu" if device < 0 else f"cuda:{device}")
+            )
         self.binary_output = binary_output
 
         # Special handling
-        if self.framework == "pt" and self.device.type == "cuda":
+        if self.framework == "pt" and self.device.type != "cpu":
             self.model = self.model.to(self.device)
 
         # Update config with task specific parameters