model

superduperdb.ext.transformers.model

Source code

LLM

LLM(self,
     db: dataclasses.InitVar[typing.Optional[ForwardRef('Datalayer')]] = None,
     uuid: str = <factory>,
     *,
     trainer: 't.Optional[Trainer]' = None,
     identifier: str = '',
     artifacts: 'dc.InitVar[t.Optional[t.Dict]]' = None,
     datatype: 'EncoderArg' = None,
     output_schema: 't.Optional[Schema]' = None,
     flatten: 'bool' = False,
     model_update_kwargs: 't.Dict' = <factory>,
     predict_kwargs: 't.Dict' = <factory>,
     compute_kwargs: 't.Dict' = <factory>,
     validation: 't.Optional[Validation]' = None,
     metric_values: 't.Dict' = <factory>,
     prompt: str = '{input}',
     prompt_func: Optional[Callable] = None,
     max_batch_size: Optional[int] = 4,
     model_name_or_path: Optional[str] = None,
     adapter_id: Union[str,
     superduperdb.ext.transformers.training.Checkpoint,
     NoneType] = None,
     model_kwargs: Dict = <factory>,
     tokenizer_kwargs: Dict = <factory>,
     prompt_template: str = '{input}') -> None

Parameter	Description
identifier	model identifier
db	Datalayer instance.
uuid	UUID of the leaf.
artifacts	A dictionary of artifacts paths and DataType objects
signature	Model signature.
datatype	DataType instance.
output_schema	Output schema (mapping of encoders).
flatten	Flatten the model outputs.
model_update_kwargs	The kwargs to use for model update.
predict_kwargs	Additional arguments to use at prediction time.
compute_kwargs	Kwargs used for compute backend job submit. Example (Ray backend): compute_kwargs = dict(resources=...).
validation	The validation Dataset instances to use.
metric_values	The metrics to evaluate on.
prompt	The template to use for the prompt.
prompt_func	prompt function, default is None
max_batch_size	The maximum batch size to use for batch generation.
model_name_or_path	model name or path
adapter_id	adapter id, default is None Add a adapter to the base model for inference.
model_kwargs	model kwargs, all the kwargs will pass to transformers.AutoModelForCausalLM.from_pretrained
tokenizer_kwargs	tokenizer kwargs, all the kwargs will pass to transformers.AutoTokenizer.from_pretrained
prompt_template	prompt template, default is "{input}"

LLM model based on transformers library.

All the model_kwargs will pass to transformers.AutoModelForCausalLM.from_pretrained. All the tokenize_kwargs will pass to transformers.AutoTokenizer.from_pretrained. When model_name_or_path, bits, model_kwargs, tokenizer_kwargs are the same, will share the same base model and tokenizer cache.

TextClassificationPipeline

TextClassificationPipeline(self,
     identifier: str,
     db: dataclasses.InitVar[typing.Optional[ForwardRef('Datalayer')]] = None,
     uuid: str = <factory>,
     *,
     preferred_devices: 't.Sequence[str]' = ('cuda',
     'mps',
     'cpu'),
     device: 't.Optional[str]' = None,
     trainer: 't.Optional[Trainer]' = None,
     artifacts: 'dc.InitVar[t.Optional[t.Dict]]' = None,
     signature: Literal['*args',
     '**kwargs',
     '*args,
    **kwargs',
     'singleton'] = 'singleton',
     datatype: 'EncoderArg' = None,
     output_schema: 't.Optional[Schema]' = None,
     flatten: 'bool' = False,
     model_update_kwargs: 't.Dict' = <factory>,
     predict_kwargs: 't.Dict' = <factory>,
     compute_kwargs: 't.Dict' = <factory>,
     validation: 't.Optional[Validation]' = None,
     metric_values: 't.Dict' = <factory>,
     tokenizer_name: Optional[str] = None,
     tokenizer_cls: object = <class 'transformers.models.auto.tokenization_auto.AutoTokenizer'>,
     tokenizer_kwargs: Dict = <factory>,
     model_name: Optional[str] = None,
     model_cls: object = <class 'transformers.models.auto.modeling_auto.AutoModelForSequenceClassification'>,
     model_kwargs: Dict = <factory>,
     pipeline: Optional[transformers.pipelines.base.Pipeline] = None,
     task: str = 'text-classification') -> None

Parameter	Description
identifier	Identifier of the leaf.
db	Datalayer instance.
uuid	UUID of the leaf.
artifacts	A dictionary of artifacts paths and DataType objects
signature	Model signature.
datatype	DataType instance.
output_schema	Output schema (mapping of encoders).
flatten	Flatten the model outputs.
model_update_kwargs	The kwargs to use for model update.
predict_kwargs	Additional arguments to use at prediction time.
compute_kwargs	Kwargs used for compute backend job submit. Example (Ray backend): compute_kwargs = dict(resources=...).
validation	The validation Dataset instances to use.
metric_values	The metrics to evaluate on.
tokenizer_name	tokenizer name
tokenizer_cls	tokenizer class, e.g. transformers.AutoTokenizer
tokenizer_kwargs	tokenizer kwargs, will pass to tokenizer_cls
model_name	model name, will pass to model_cls
model_cls	model class, e.g. AutoModelForSequenceClassification
model_kwargs	model kwargs, will pass to model_cls
pipeline	pipeline instance, default is None, will build when None
task	task of the pipeline
trainer	TransformersTrainer instance
preferred_devices	preferred devices
device	device to use

A wrapper for transformers.Pipeline.

# Example:
# -------
model = TextClassificationPipeline(...)