Description #

Text embedding generation processor

Key features #

• Text embedding support: Supports generating embeddings for text, enabling text retrieval capabilities.

Input and output specifications: #

• Input format:
  • Text data: string type
• Output format:
  • Default mode: vector representation as a float array

Operator parameters #

Input #

Output #

Returns the embedding array processed by the model. Type: list[float]

Parameters #

If a parameter does not have a default value, it is required.

Examples #

The following code demonstrates how to use daft to access the ModelArk text embedding model for batch inference. Note that the results of each large model inference may vary.

from __future__ import annotations

import os

import daft
from daft import col
from daft.las.functions.ark_llm.doubao_embedding_text import DoubaoEmbeddingText
from daft.las.functions.udf import las_udf

if os.getenv("DAFT_RUNNER", "native") == "ray":
    import logging

    import ray

    def configure_logging():
        logging.basicConfig(
            level=logging.INFO,
            format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
            datefmt="%Y-%m-%d %H:%M:%S.%s".format(),
        )
        logging.getLogger("tracing.span").setLevel(logging.WARNING)
        logging.getLogger("daft_io.stats").setLevel(logging.WARNING)
        logging.getLogger("DaftStatisticsManager").setLevel(logging.WARNING)
        logging.getLogger("DaftFlotillaScheduler").setLevel(logging.WARNING)
        logging.getLogger("DaftFlotillaDispatcher").setLevel(logging.WARNING)

    ray.init(dashboard_host="0.0.0.0", runtime_env={"worker_process_setup_hook": configure_logging})
    daft.set_runner_ray()

daft.set_execution_config(min_cpu_per_task=0)

if __name__ == "__main__":
    # You need to configure the environment variable LAS_API_KEY: Obtain LAS_API_KEY by creating it on the LAS service page
    samples = {"text": ["Hello World!", None]}

    df = daft.from_pydict(samples)
    # Compute the embedding data for text
    df = df.with_column(
        "embeddings",
        las_udf(
            DoubaoEmbeddingText,
            construct_args={
                "model": "doubao-embedding-large",
            },
        )(col("text")),
    )

    df.show()

    #  Output (the results of each large model inference may vary)
    # ╭────────┬───────────────────╮
    # │ text         ┆ embeddings                     │
    # │ ---          ┆ ---                            │
    # │ Utf8         ┆ List[Float32]                  │
    # ╞════════╪═══════════════════╡
    # │ Hello World! ┆ [0.080078125, 1.8359375, 0.84… │
    # ├╌╌╌╌╌╌╌╌╌╌╌╌╌╌┼╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌╌┤
    # │ None         ┆ None                           │
    # ╰────────┴───────────────────╯