StypeConfig

class sdm.stype.StypeConfig(categorical: CategoricalConfig = <factory>, text: TextConfig = <factory>)

Semantic type inference thresholds.

Parameters:

• categorical (CategoricalConfig) – Thresholds for inferring numeric columns as categorical.

• text (TextConfig) – Thresholds for inferring string columns as text.

CategoricalConfig

class sdm.stype.CategoricalConfig(min_rows: int = 151, max_unique: int = 3)

Thresholds for inferring numeric columns as categorical.

Parameters:

• min_rows (int) – Minimum row count, including nulls.

• max_unique (int) – Maximum distinct value count, including nulls.

TextConfig

class sdm.stype.TextConfig(min_unique: int = 200, min_unique_ratio: float = 0.05, min_avg_words: float = 3)

Thresholds for inferring string columns as text.

Parameters:

• min_unique (int) – Minimum number of distinct strings, excluding nulls.

• min_unique_ratio (float) – Minimum ratio of distinct strings, excluding nulls.

• min_avg_words (float) – Minimum average number of words per distinct string.