Class AlloyDBToolSpec (0.4.0)

AlloyDBToolSpec(
    engine: llama_index_alloydb_pg.engine.AlloyDBEngine,
    model_id: typing.Optional[str] = None,
)

Tools that run AlloyDB AI functions in the database.

AlloyDBToolSpec(engine).to_tool_list() returns one FunctionTool per function, each with a native async path, ready for an agent such as FunctionAgent. Requires AlloyDB for PostgreSQL 17 or later with the google_ml_integration extension 1.5.7 or later and the google_ml_integration.enable_ai_query_engine and google_ml_integration.enable_preview_ai_functions flags on. See https://cloud.google.com/alloydb/docs/ai/analyze-sentiment.

Parameters
Name	Description
`engine`	`AlloyDBEngine` Connection pool to the AlloyDB database.
`model_id`	`Optional[str]` ID of the model the AI functions use. If None, the database uses the model named by its `google_ml_integration.default_llm_model` flag. Defaults to None.

Methods

aanalyze_sentiment

aanalyze_sentiment(text: str) -> str

Classify the sentiment of a text as positive, negative or neutral.

Parameter
Name	Description
`text`	`str` The text to analyze, such as a review or a message.

Exceptions
Type	Description
`ValueError`	If `text` is blank or the model gives no answer.

Returns
Type	Description
`str`	"positive", "negative" or "neutral".

aevaluate_condition

aevaluate_condition(condition: str) -> bool

Decide whether a natural-language condition is true or false.

Parameter
Name	Description
`condition`	`str` A yes or no question or a statement that includes the text it is about, for example "Does the review 'The battery died in a day' complain about the battery?".

Exceptions
Type	Description
`ValueError`	If `condition` is blank or the model gives no answer.

Returns
Type	Description
`bool`	True if the condition holds, otherwise False.

analyze_sentiment

analyze_sentiment(text: str) -> str

Classify the sentiment of a text as positive, negative or neutral.

Parameter
Name	Description
`text`	`str` The text to analyze, such as a review or a message.

Exceptions
Type	Description
`ValueError`	If `text` is blank or the model gives no answer.

Returns
Type	Description
`str`	"positive", "negative" or "neutral".

asummarize

asummarize(text: str, *, instructions: typing.Optional[str] = None) -> str

Summarize a text, such as an article or a set of reviews.

Parameters
Name	Description
`text`	`str` The text to summarize.
`instructions`	`Optional[str]` Extra instructions for the summary, for example "Use three bullet points". Defaults to None.

Exceptions
Type	Description
`ValueError`	If `text` is blank or the model gives no answer.

Returns
Type	Description
`str`	The summary, as the model returned it. It is often wrapped in double quotes.

evaluate_condition

evaluate_condition(condition: str) -> bool