Error:缺少 NLTK 资源 punkt_tab

[oklbuy331]

1️⃣ 描述一下问题

使用bge-m3，采用Semantic分块策略对文档进行解析处理时，提示文档入库失败。

2️⃣ 报错日志

请运行以下命令，并提供部分相关日志：

# macOS / Linux
make logs

# Windows
docker logs --tail=100 api-dev
git rev-parse HEAD

make logs 的输出：
07-03 13:24:06 WARNING base.py:1242: File file_e58621 has indexing status but is not in processing queue, marking as error_indexing
07-03 13:24:06 INFO base.py:1255: Fixed interrupted processing status for database kb_hbqb9xl458
07-03 13:24:06 DEBUG milvus.py:664: [index_file] file_id=file_e58621, processing_params={'ocr_engine': 'rapid_ocr', 'ocr_engine_config': {}, 'chunk_parser_config': {'overlapped_percent': 14}, 'content_type': 'file', 'chunk_preset_id': 'semantic', 'chunk_engine_version': 'ragflow_like_v1', 'image_bucket': 'public', 'image_prefix': 'kb_hbqb9xl458/kb-images'}
07-03 13:24:06 DEBUG base.py:1188: Added file file_e58621 to processing queue
07-03 13:24:06 DEBUG kb_utils.py:240: Parsed MinIO URL: bucket_name=knowledgebases, object_name=kb_hbqb9xl458/parsed/file_e58621.md
07-03 13:24:06 INFO client.py:217: 成功下载 'kb_hbqb9xl458/parsed/file_e58621.md' 从存储桶 'knowledgebases'
07-03 13:24:06 INFO semantic.py:135: 语义切分开始: max_length=512, content_length=56164
07-03 13:24:06 INFO semantic.py:144: 语义切分加载Embedding模型: siliconflow-cn:Pro/BAAI/bge-m3
07-03 13:24:06 INFO embed.py:258: Selecting embedding model: siliconflow-cn:Pro/BAAI/bge-m3 (provider_type=openai)
07-03 13:24:06 INFO __init__.py:64: Running in docker, using https://api.siliconflow.cn/v1/embeddings as base url
07-03 13:24:06 ERROR milvus.py:699: Indexing failed for file_e58621: 缺少 NLTK 资源 punkt_tab。请先执行: python -m nltk.downloader punkt_tab
07-03 13:24:06 DEBUG base.py:1200: Removed file file_e58621 from processing queue
07-03 13:24:06 ERROR knowledge_router.py:628: 自动入库失败 http://localhost:9000/knowledgebases/kb_hbqb9xl458/upload/水利工程安全监测智能分析预警平台开发及应用部署__招标文件_技术标准和要求_1783056242841.docx (file_id=file_e58621): 缺少 NLTK 资源 punkt_tab。请先执行: python -m nltk.downloader punkt_tab

3️⃣ 相关截图

#️⃣ 其他相关信息

✅ 如果问题与模型调用相关，请尝试切换到其他在线模型

[xerrors]

docker exec api-dev python -m nltk.downloader punkt_tab
docker exec worker-dev python -m nltk.downloader punkt_tab

尝试一下