mindformers.pipeline.pipeline

mindformers.pipeline.pipeline(task: str = None, model: Union[str, mindformers.models.base_model.BaseModel, mindspore.train.model.Model, None] = None, tokenizer: Optional[mindformers.models.base_tokenizer.BaseTokenizer] = None, image_processor: Optional[mindformers.models.base_processor.BaseImageProcessor] = None, audio_processor: Optional[mindformers.models.base_processor.BaseAudioProcessor] = None, **kwargs)

Pipeline for downstream tasks

参数

• task (str) – The supported task could be selected from MindFormerBook.show_pipeline_support_task_list().

• model (Optional[Union[str, BaseModel]]) – The model used for task.

• tokenizer (Optional[BaseTokenizer]) – The tokenizer of the model.

• image_processor (Optional[BaseImageProcessor]) – The image processor of the model.

• audio_processor (Optional[BaseAudioProcessor]) – The audio processor of the model.

返回

A task pipeline.

引发

KeyError – If the task or model is not supported.

实际案例

>>> from mindformers import pipeline
>>> from mindformers.tools.image_tools import load_image
>>> classifier = pipeline("zero_shot_image_classification",
    candidate_labels=["sunflower", "tree", "dog", "cat", "toy"])
>>> img = load_image("https://ascend-repo-modelzoo.obs.cn-east-2."
    "myhuaweicloud.com/XFormer_for_mindspore/clip/sunflower.png")
>>> classifier(img)
    [[{'score': 0.99995565, 'label': 'sunflower'},
    {'score': 2.5318595e-05, 'label': 'toy'},
    {'score': 9.903885e-06, 'label': 'dog'},
    {'score': 6.75336e-06, 'label': 'tree'},
    {'score': 2.396818e-06, 'label': 'cat'}]]