Runner#

class FlagEmbedding.finetune.embedder.decoder_only.base.DecoderOnlyEmbedderRunner(model_args: DecoderOnlyEmbedderModelArguments, data_args: AbsEmbedderDataArguments, training_args: AbsEmbedderTrainingArguments)[source]#

Runner class for decoder only embedding model.

Parameters:
load_tokenizer_and_model() Tuple[PreTrainedTokenizer, AbsEmbedderModel][source]#

Load tokenizer and model.

Returns:

Tokenizer and model instances.

Return type:

Tuple[PreTrainedTokenizer, AbsEmbedderModel]

load_trainer() DecoderOnlyEmbedderTrainer[source]#

Load the trainer.

Returns:

Loaded trainer instance.

Return type:

DecoderOnlyEmbedderTrainer

run()[source]#

Run the finetune.