From 04528cb2921b236c95588ff611b3e16efa0dba9c Mon Sep 17 00:00:00 2001
From: 嘉渊 <wangjiaming.wjm@alibaba-inc.com>
Date: 星期四, 15 六月 2023 17:20:07 +0800
Subject: [PATCH] update repo
---
funasr/build_utils/build_streaming_iterator.py | 16 +++++++++++++---
1 files changed, 13 insertions(+), 3 deletions(-)
diff --git a/funasr/build_utils/build_streaming_iterator.py b/funasr/build_utils/build_streaming_iterator.py
index 57cf8cf..8c5f7fc 100644
--- a/funasr/build_utils/build_streaming_iterator.py
+++ b/funasr/build_utils/build_streaming_iterator.py
@@ -5,7 +5,7 @@
from funasr.datasets.iterable_dataset import IterableESPnetDataset
from funasr.datasets.small_datasets.collate_fn import CommonCollateFn
from funasr.datasets.small_datasets.preprocessor import build_preprocess
-
+from funasr.build_utils.build_model_from_file import build_model_from_file
def build_streaming_iterator(
task_name,
@@ -17,6 +17,8 @@
mc: bool = False,
dtype: str = np.float32,
num_workers: int = 1,
+ use_collate_fn: bool = True,
+ preprocess_fn=None,
ngpu: int = 0,
train: bool=False,
) -> DataLoader:
@@ -24,10 +26,18 @@
assert check_argument_types()
# preprocess
- preprocess_fn = build_preprocess(preprocess_args, train)
+ if preprocess_fn is not None:
+ preprocess_fn = preprocess_fn
+ elif preprocess_args is not None:
+ preprocess_args.task_name = task_name
+ preprocess_fn = build_preprocess(preprocess_args, train)
+ else:
+ preprocess_fn = None
# collate
- if task_name in ["punc", "lm"]:
+ if not use_collate_fn:
+ collate_fn = None
+ elif task_name in ["punc", "lm"]:
collate_fn = CommonCollateFn(int_pad_value=0)
else:
collate_fn = CommonCollateFn(float_pad_value=0.0, int_pad_value=-1)
--
Gitblit v1.9.1