From e9acc5db07daa51a22cd51ea9233ee09a38d726d Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期四, 06 六月 2024 18:36:22 +0800
Subject: [PATCH] auto frontend
---
funasr/datasets/openai_datasets/datasets.py | 11 +++++------
1 files changed, 5 insertions(+), 6 deletions(-)
diff --git a/funasr/datasets/openai_datasets/datasets.py b/funasr/datasets/openai_datasets/datasets.py
index 9a542ad..9bd0698 100644
--- a/funasr/datasets/openai_datasets/datasets.py
+++ b/funasr/datasets/openai_datasets/datasets.py
@@ -123,21 +123,20 @@
) # speech: [b, T, d]
if self.permute:
speech = speech.permute(0, 2, 1)
- if speech_lengths > self.batch_size:
- continue
+ # if speech_lengths > self.batch_size:
+ # continue
- fbank_lens = speech_lengths[0].item()
- olens = 1 + (fbanks_len - 3 + 2 * 1) // 2
+ olens = 1 + (speech_lengths[0].item() - 3 + 2 * 1) // 2
olens = 1 + (olens - 3 + 2 * 1) // 2
sub_token_len = (olens - 1) // 2 + 1
- sub_token = [0] * sub_token_len[0]
+ sub_token = [0] * sub_token_len
fbank_beg_i = [len(source_ids)]
source_ids += sub_token
fbank_mask_i += [1] * len(sub_token)
source_mask = [-100] * len(source_ids)
target_out = f"{target_out}<|im_end|>"
- target_ids = tokenizer.encode(target_out)
+ target_ids = self.tokenizer.encode(target_out)
input_ids += source_ids + target_ids
labels += source_mask + target_ids
fbank_mask += fbank_mask_i
--
Gitblit v1.9.1