Function preprocess_function_train_problem_solving

codes/main.py:327–393

(examples)

        return model_inputs

    def preprocess_function_train_problem_solving(examples):
        max_seq_length = data_args.max_source_length + data_args.max_target_length + 1

        model_inputs = {
            "input_ids": [],
            "labels": [],
        }
        for i in range(len(examples[prompt_column])):
            if examples[prompt_column][i] and examples[response_column][i]:
                query, answer = examples[prompt_column][i], examples[response_column][i]

                history = (
                    examples[history_column][i] if history_column is not None else None
                )
                # pre_prompt = "Please analyze and solve the following problem step by step: "
                # prompt = " Let's think step by step."
                # prompt = tokenizer.build_single_message("user", "", message=prefix + query + prompt)
                # prompt = [tokenizer.get_command("<|user|>")] + tokenizer.encode(
                #     text=prefix + query,
                #     add_special_tokens=False,
                #     truncation=True,
                #     max_length=data_args.max_source_length,
                # )
                prompt = tokenizer.build_single_message("user", "", prefix + query)
                prompt += [tokenizer.get_command("<|assistant|>")]
                # prompt = tokenizer.build_prompt(query, history)
                a_ids = prompt
                b_ids = tokenizer.encode(
                    text=answer,
                    add_special_tokens=False,
                    truncation=True,
                    max_length=data_args.max_target_length,
                )

                # a_ids = tokenizer.encode(
                #     text=prompt,
                #     add_special_tokens=True,
                #     truncation=True,
                #     max_length=data_args.max_source_length,
                # )
                # b_ids = tokenizer.encode(
                #     text=answer,
                #     add_special_tokens=False,
                #     truncation=True,
                #     max_length=data_args.max_target_length,
                # )

                context_length = len(a_ids)
                input_ids = a_ids + b_ids + [tokenizer.eos_token_id]
                labels = (
                    [tokenizer.pad_token_id] * context_length
                    + b_ids
                    + [tokenizer.eos_token_id]
                )

                pad_len = max_seq_length - len(input_ids)
                input_ids = input_ids + [tokenizer.pad_token_id] * pad_len
                labels = labels + [tokenizer.pad_token_id] * pad_len