dataquality.loggers.logger_config.seq2seq package

Submodules

dataquality.loggers.logger_config.seq2seq.chat module

pydantic model Seq2SeqChatLoggerConfig

Bases: Seq2SeqLoggerConfig

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Fields:

dataquality.loggers.logger_config.seq2seq.completion module

pydantic model Seq2SeqCompletionLoggerConfig

Bases: Seq2SeqLoggerConfig

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Fields:

dataquality.loggers.logger_config.seq2seq.seq2seq_base module

pydantic model Seq2SeqLoggerConfig

Bases: BaseLoggerConfig

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Fields:

• generation_config (transformers.generation.configuration_utils.GenerationConfig | None)

• generation_splits (Set[dataquality.schemas.split.Split])

• id_to_formatted_prompt_length (Dict[str, Dict[int, int]])

• id_to_tokens (Dict[str, Dict[int, List[int]]])

• max_input_tokens (int | None)

• max_target_tokens (int | None)

• model (transformers.modeling_utils.PreTrainedModel | peft.peft_model.PeftModel | None)

• model_type (dataquality.schemas.seq2seq.Seq2SeqModelType | None)

• response_template (List[int] | None)

• sample_length (Dict[str, int])

• tokenizer (transformers.tokenization_utils_fast.PreTrainedTokenizerFast | None)

field generation_config: Optional[GenerationConfig] = None

field generation_splits: Set[Split] = {}

field id_to_formatted_prompt_length: Dict[str, Dict[int, int]] = {}

field id_to_tokens: Dict[str, Dict[int, List[int]]] = {}

field max_input_tokens: Optional[int] = None

field max_target_tokens: Optional[int] = None

field model: Union[PreTrainedModel, PeftModel, None] = None

field model_type: Optional[Seq2SeqModelType] = None

field response_template: Optional[List[int]] = None

field sample_length: Dict[str, int] = {}

field tokenizer: Optional[PreTrainedTokenizerFast] = None

Module contents