AbsTrainer

AbsEmbedderTrainer

class FlagEmbedding.abc.finetune.embedder.AbsEmbedderTrainer(*args, processing_class=None, tokenizer=None, **kwargs)

Abstract class for the trainer of embedder.

Methods

AbsEmbedderTrainer.compute_loss(model, inputs, return_outputs=False, **kwargs)

How the loss is computed by Trainer. By default, all models return the loss in the first element.

Subclass and override for custom behavior.

Parameters:

• model (AbsEmbedderModel) – The model being trained.

• inputs (dict) – A dictionary of input tensors to be passed to the model.

• return_outputs (bool, optional) – If True, returns both the loss and the model’s outputs. Otherwise, returns only the loss.

Returns:

The computed loss. If return_outputs is True,

also returns the model’s outputs in a tuple (loss, outputs).

Return type:

Union[torch.Tensor, tuple(torch.Tensor, EmbedderOutput)]