Are you an LLM? You can read better optimized documentation at /api/@nhtio/adk/batteries/embeddings/transformers_js/types/type-aliases/CreateTransformersJsEmbeddingsPipeline.md for this page in Markdown format
Type Alias: CreateTransformersJsEmbeddingsPipeline
ts
type CreateTransformersJsEmbeddingsPipeline = (input: {
device?: TransformersJsEmbeddingsDeviceType;
dtype?: TransformersJsEmbeddingsDataType;
model: string;
onInitProgress?: TransformersJsEmbeddingsProgressCallback;
}) => Promise<TransformersJsEmbeddingsPipeline>;Defined in: src/batteries/embeddings/transformers_js/types.ts:81
Factory for lazily creating a feature-extraction pipeline. Defaults to a dynamic import of @huggingface/transformers + pipeline('feature-extraction', …); override to inject a pre-built pipeline or a test double.
Parameters
| Parameter | Type |
|---|---|
input | { device?: TransformersJsEmbeddingsDeviceType; dtype?: TransformersJsEmbeddingsDataType; model: string; onInitProgress?: TransformersJsEmbeddingsProgressCallback; } |
input.device? | TransformersJsEmbeddingsDeviceType |
input.dtype? | TransformersJsEmbeddingsDataType |
input.model | string |
input.onInitProgress? | TransformersJsEmbeddingsProgressCallback |
Returns
Promise<TransformersJsEmbeddingsPipeline>