fix bug

游雁

2024-06-09 1186cd96a5a8fa3466c5e3e41f86f0280deb1410

 funasr/models/llm_asr/model.py

@@ -6,7 +6,7 @@
import torch.nn as nn
import torch.nn.functional as F
from torch.cuda.amp import autocast

import re
from funasr.models.scama.utils import sequence_mask
from funasr.losses.label_smoothing_loss import LabelSmoothingLoss
from funasr.models.ctc.ctc import CTC
@@ -19,6 +19,7 @@
from funasr.utils.datadir_writer import DatadirWriter
from funasr.register import tables
from funasr.train_utils.device_funcs import to_device
import traceback


@tables.register("model_classes", "LLMASR")
@@ -489,6 +490,7 @@
            fbank_fake_len = fbank_fake_lens[batch_idx].item()
            fbank_beg_idx = fbank_beg[batch_idx, 0].item()
            min_len = min(fbank_fake_len, inputs_embeds.shape[1] - fbank_beg_idx)

            try:
                inputs_embeds[batch_idx, fbank_beg_idx : fbank_beg_idx + min_len, :] = encoder_out[
                    batch_idx, :min_len, :
@@ -496,10 +498,10 @@
            except Exception as e:
                logging.error(f"{str(e)}, {traceback.format_exc()}")
                logging.info(
                    f"batch_idx: {batch_idx}, inputs_embeds: {inputs_embeds.shape}, fbank_beg_idx: {fbank_beg_idx}, min_len: {min_len}, fbank_fake_len: {fbank_fake_len}"
                    f"batch_idx: {batch_idx}, inputs_embeds: {inputs_embeds.shape}, fbank_beg_idx: {fbank_beg_idx}, min_len: {min_len}, fbank_fake_len: {fbank_fake_len}, encoder_out: {encoder_out.shape}, encoder_out_lens: {encoder_out_lens[batch_idx].item()}"
                )
                fbank_fake_len = encoder_out_lens[batch_idx].item()
                min_len = min(fbank_fake_len, inputs_embeds.shape[1] - fbank_beg_idx)
                min_len = min(fbank_fake_len, min_len)
                inputs_embeds[batch_idx, fbank_beg_idx : fbank_beg_idx + min_len, :] = encoder_out[
                    batch_idx, :min_len, :
                ]
@@ -532,7 +534,7 @@
        loss, stats, weight = force_gatherable((loss, stats, batch_size), loss.device)
        return loss, stats, weight

    def data_template(self, data_in):
    def data_template(self, data):
        system, user, assistant = [], [], []
        for i, item in enumerate(data):
            role = item["role"]
@@ -560,21 +562,31 @@
        user = contents["user"]
        assistant = contents["assistant"]
        pattern = re.compile(r"(<\|startofspeech\|>.*?<\|endofspeech\|>)")
        input_ids, labels, fbank, fbank_lens, fbank_mask, fbank_beg = [], [], [], [], [], []
        input_ids, labels, source_ids, target_ids, fbank, fbank_lens, fbank_mask, fbank_beg = (
            [],
            [],
            [],
            [],
            [],
            [],
            [],
            [],
        )

        for i, (system_prompt, user_prompt, target_out) in enumerate(zip(system, user, assistant)):

            source_input = f"<|im_start|>system\n{system_prompt}<|im_end|>\n<|im_start|>user\n{user_prompt}<|im_end|>\n<|im_start|>assistant\n"

            splits = pattern.split(source_input)
            source_ids = []
            source_ids_i = []
            fbank_mask_i = []
            fbank_beg_i = []
            fbank_lens_i = []
            # target_ids_i = []
            for k, sub_str in enumerate(splits):
                if not sub_str.startswith("<|startofspeech|>"):
                    sub_token = tokenizer.encode(sub_str)
                    source_ids += sub_token
                    source_ids_i += sub_token
                    fbank_mask_i += [0] * len(sub_token)
                else:
                    sub_str = sub_str.replace("<|startofspeech|>", "").replace(
@@ -600,14 +612,14 @@
                        olens = 1 + (olens - 3 + 2 * 1) // 2
                        sub_token_len = (olens - 1) // 2 + 1
                        sub_token = [0] * sub_token_len
                        fbank_beg_i = [len(source_ids)]
                        source_ids += sub_token
                        fbank_beg_i = [len(source_ids_i)]
                        source_ids_i += sub_token
                        fbank_mask_i += [1] * len(sub_token)

            source_mask = [-100] * len(source_ids)
            source_mask = [-100] * len(source_ids_i)
            target_out = f"{target_out}<|im_end|>"
            target_ids = tokenizer.encode(target_out)
            input_ids += source_ids + target_ids
            input_ids += source_ids_i + target_ids
            labels += source_mask + target_ids
            fbank_mask += fbank_mask_i
            fbank_beg.append(fbank_beg_i)
@@ -615,7 +627,7 @@
        input_ids = torch.tensor(input_ids, dtype=torch.int64)  # [: self.max_token_length]
        attention_mask = torch.tensor([1] * len(input_ids), dtype=torch.int32)
        labels = torch.tensor(labels, dtype=torch.int64)  # [: self.max_token_length]
        source_ids = torch.tensor(source_ids, dtype=torch.int64)
        source_ids = torch.tensor(source_ids_i, dtype=torch.int64)
        target_ids = torch.tensor(target_ids, dtype=torch.int64)

        fbank = speech[0, :, :]
@@ -653,7 +665,7 @@
        if kwargs.get("batch_size", 1) > 1:
            raise NotImplementedError("batch decoding is not implemented")

        contents = self.data_template(data_in)
        contents = self.data_template(data_in[0])
        output = self.data_load_speech(contents, tokenizer, frontend, **kwargs)
        batch = to_device(output, kwargs["device"])

@@ -667,7 +679,7 @@

        input_ids = batch["input_ids"]
        source_ids = batch["source_ids"]
        if kwargs.get("tearchforing", False):
        if not kwargs.get("tearchforing", False):
            input_ids = source_ids
        input_ids[input_ids < 0] = 0
        inputs_embeds = self.llm.model.get_input_embeddings()(input_ids)
@@ -682,6 +694,7 @@
                batch_idx, :min_len, :
            ]

        label = contents["assistant"][0]
        if not kwargs.get("tearchforing", False):

            generated_ids = self.llm.generate(
@@ -694,7 +707,24 @@
            response = tokenizer.batch_decode(
                generated_ids, skip_special_tokens=kwargs.get("skip_special_tokens", True)
            )[0]
            label = contents["assistant"][0]

            loss = None
        else:

            labels_ids = batch["labels_ids"]
            labels_ids[labels_ids == -1] = -100
            attention_mask = batch.get("attention_mask", None)
            model_outputs = self.llm(
                inputs_embeds=inputs_embeds, attention_mask=attention_mask, labels=labels_ids
            )

            preds = torch.argmax(model_outputs.logits, -1)[:, source_ids.shape[1] :]
            response = tokenizer.batch_decode(
                preds,
                add_special_tokens=False,
                skip_special_tokens=kwargs.get("skip_special_tokens", True),
            )[0]
            loss = model_outputs.loss.item()

        ibest_writer = None
        if kwargs.get("output_dir") is not None:
@@ -704,10 +734,12 @@

        results = []
        result_i = {"key": key[0], "text": response, "label": label}
        if loss is not None:
            result_i["loss"] = loss
        results.append(result_i)

        if ibest_writer is not None:
            ibest_writer["text"][key[0]] = text
            ibest_writer["text"][key[0]] = response
            ibest_writer["label"][key[0]] = label

        return results, meta_data