Database and task specifications

Looking for usage documentation? Check out Relational predictions.

`DatabaseSpec`

A relational database: its named table sources and their schema. Carries no val/test timestamps — those define an evaluation split, not the data, and enter at build_dataset / UserDataset (mirroring RelBench’s Database vs Dataset split).

DatabaseSpec(
    tables: dict[str, TableSource],
)

Fields

Name	Type	Default	Description
`tables`	`dict[str, TableSource]`	Required	—

`DatabaseSpec.from_yaml`

View source

Load a database schema from a YAML file (table name -> its schema). Each table’s data file is its path if given; relative paths and the default <name>.parquet resolve against data_dir, or the YAML’s own directory if data_dir is omitted - so the schema stays portable and materialize_relbench can supply the files.

DatabaseSpec.from_yaml(
    path: str,
    *,
    data_dir: str | None = None,
) -> DatabaseSpec

Parameters

Parameter	Type	Default	Description
`path`	`str`	Required	—
`data_dir`	`str \| None`	`None`	—

Returns

Type	Description
`DatabaseSpec`	—

`PredictiveTaskSpec`

View source

One training task: label SQL and temporal evaluation splits. The query is a DuckDB statement that joins timestamp_df (a single column timestamp holding the anchor times) against the database tables - referenced by their names - and returns exactly time_col, entity_col and target_col, computing the label over the forward window (t, t + timedelta]. The literal {timedelta} in the query is replaced with the string form of timedelta, e.g. INTERVAL "{timedelta}". val_timestamp / test_timestamp are the tuning and final split cutoffs. The database schema lives separately in a DatabaseSpec. task_type accepts the RelBench enum or the strings "binary_classification" / "regression"; timedelta accepts a pandas.Timedelta or anything it can parse.

PredictiveTaskSpec(
    entity_table: str,
    entity_col: str,
    time_col: str,
    target_col: str,
    task_type: TaskType,
    timedelta: pd.Timedelta,
    query: str,
    val_timestamp: pd.Timestamp,
    test_timestamp: pd.Timestamp,
    num_eval_timestamps: int = 1,
)

Fields

Name	Type	Default	Description
`entity_table`	`str`	Required	—
`entity_col`	`str`	Required	—
`time_col`	`str`	Required	—
`target_col`	`str`	Required	—
`task_type`	`TaskType`	Required	—
`timedelta`	`pd.Timedelta`	Required	—
`query`	`str`	Required	—
`val_timestamp`	`pd.Timestamp`	Required	—
`test_timestamp`	`pd.Timestamp`	Required	—
`num_eval_timestamps`	`int`	`1`	—

Reference

tabpfn

tabpfn-client

tabpfn-extensions

tabpfn-rel

REST API

Database and task specifications

`DatabaseSpec`

`DatabaseSpec.from_yaml`

`PredictiveTaskSpec`

​DatabaseSpec

​DatabaseSpec.from_yaml

​PredictiveTaskSpec

`DatabaseSpec`

`DatabaseSpec.from_yaml`

`PredictiveTaskSpec`