Files
dsProject/dsLightRag/T1_Train.py

70 lines
2.3 KiB
Python
Raw Normal View History

2025-08-14 15:45:08 +08:00
import asyncio
import logging
import os
from Util.DocxUtil import get_docx_content_by_pandoc
from Util.LightRagUtil import initialize_rag
# 是不是清空重新生成
IS_CLEAR = False
# 更详细地控制日志输出
logger = logging.getLogger('lightrag')
logger.setLevel(logging.INFO)
handler = logging.StreamHandler()
handler.setFormatter(logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s'))
logger.addHandler(handler)
async def main():
# 清空文件
if IS_CLEAR:
# 注释掉或删除以下清理代码
files_to_delete = [
"graph_chunk_entity_relation.graphml",
"kv_store_doc_status.json",
"kv_store_full_docs.json",
"kv_store_text_chunks.json",
"vdb_chunks.json",
"vdb_entities.json",
"vdb_relationships.json",
]
# 删除文件
for file in files_to_delete:
file_path = os.path.join(WORKING_DIR, file)
if os.path.exists(file_path):
os.remove(file_path)
logger.info(f"删除的文件:: {file_path}")
try:
# 注意默认设置使用NetworkX
rag = await initialize_rag(WORKING_DIR)
# 在docx_file 目录下遍历所有以KEMU开头的文件
for filename in os.listdir(docx_file):
if filename.startswith(KEMU):
file_path = os.path.join(docx_file, filename)
# 获取docx文件的内容
if filename.endswith(".docx"):
continue
# content = get_docx_content_by_pandoc(file_path)
elif filename.endswith(".txt"):
with open(file_path, "r", encoding="utf-8") as f:
content = f.read()
await rag.ainsert(content, file_paths=[filename])
logger.info(f"Inserted content from {filename}")
except Exception as e:
logger.error(f"An error occurred: {e}")
finally:
await rag.finalize_storages()
# KEMUS=['JiHe','Math','SuShi','Chemistry','ShiJi','ChangChun']
KEMUS = ['Math2']
for KEMU in KEMUS:
# 组装文件路径
WORKING_DIR = "./Topic/" + KEMU
docx_file = 'static/Txt/'
asyncio.run(main())