def __call__()

in trl/trainer/utils.py [0:0]

• 52 lines of code
• 10 McCabe index (conditional complexity)

    def __call__(self, features: list[dict[str, Any]]) -> dict[str, Any]:
        features_chosen = []
        features_rejected = []
        margin = []
        # check if we have a margin. If we do, we need to batch it as well
        has_margin = "margin" in features[0]
        for feature in features:
            # check if the keys are named as expected
            if (
                "input_ids_chosen" not in feature
                or "input_ids_rejected" not in feature
                or "attention_mask_chosen" not in feature
                or "attention_mask_rejected" not in feature
            ):
                raise ValueError(
                    "The features should include `input_ids_chosen`, `attention_mask_chosen`, `input_ids_rejected` and `attention_mask_rejected`"
                )

            features_chosen.append(
                {
                    "input_ids": feature["input_ids_chosen"],
                    "attention_mask": feature["attention_mask_chosen"],
                }
            )
            features_rejected.append(
                {
                    "input_ids": feature["input_ids_rejected"],
                    "attention_mask": feature["attention_mask_rejected"],
                }
            )
            if has_margin:
                margin.append(feature["margin"])
        batch_chosen = self.tokenizer.pad(
            features_chosen,
            padding=self.padding,
            pad_to_multiple_of=self.pad_to_multiple_of,
            return_tensors=self.return_tensors,
        )
        batch_rejected = self.tokenizer.pad(
            features_rejected,
            padding=self.padding,
            pad_to_multiple_of=self.pad_to_multiple_of,
            return_tensors=self.return_tensors,
        )
        batch = {
            "input_ids_chosen": batch_chosen["input_ids"],
            "attention_mask_chosen": batch_chosen["attention_mask"],
            "input_ids_rejected": batch_rejected["input_ids"],
            "attention_mask_rejected": batch_rejected["attention_mask"],
            "return_loss": True,
        }
        if has_margin:
            margin = torch.tensor(margin, dtype=torch.float)
            batch["margin"] = margin
        return batch