Introducing LiteRT: Google's high-performance runtime for on-device AI, formerly known as TensorFlow Lite. Learn more

mp.tasks.components.containers.Embedding

Embedding result for a given embedder head.

Main aliases

mp.tasks.components.containers.Embedding(
    embedding: np.ndarray,
    head_index: Optional[int] = None,
    head_name: Optional[str] = None
)

Attributes
`embedding`	The actual embedding, either floating-point or scalar-quantized.
`head_index`	The index of the embedder head that produced this embedding. This is useful for multi-head models.
`head_name`	The name of the embedder head, which is the corresponding tensor metadata name (if any). This is useful for multi-head models.

Methods

__eq__(
    other
)

Class Variables
head_index	`None`
head_name	`None`

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2024-05-01 UTC.