From 0e622e694e6cb4459955f1e5942a7c53349ce640 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期二, 19 十二月 2023 21:58:14 +0800
Subject: [PATCH] funasr2
---
funasr/tokenizer/abs_tokenizer.py | 12 ++++++++++--
1 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/funasr/tokenizer/abs_tokenizer.py b/funasr/tokenizer/abs_tokenizer.py
index bbfd173..349ebc0 100644
--- a/funasr/tokenizer/abs_tokenizer.py
+++ b/funasr/tokenizer/abs_tokenizer.py
@@ -7,6 +7,7 @@
from typing import Iterable
from typing import List
from typing import Union
+import json
import numpy as np
@@ -27,7 +28,7 @@
):
if token_list is not None:
- if isinstance(token_list, (Path, str)):
+ if isinstance(token_list, (Path, str)) and token_list.endswith(".txt"):
token_list = Path(token_list)
self.token_list_repr = str(token_list)
self.token_list: List[str] = []
@@ -36,7 +37,14 @@
for idx, line in enumerate(f):
line = line.rstrip()
self.token_list.append(line)
-
+ elif isinstance(token_list, (Path, str)) and token_list.endswith(".json"):
+ token_list = Path(token_list)
+ self.token_list_repr = str(token_list)
+ self.token_list: List[str] = []
+
+ with open('data.json', 'r', encoding='utf-8') as f:
+ self.token_list = json.loads(f.read())
+
else:
self.token_list: List[str] = list(token_list)
self.token_list_repr = ""
--
Gitblit v1.9.1