[improvement][chat] Move python code out of chat-core module

chat/python/instances/chromadb_instance.py

@@ -0,0 +1,21 @@
+# -*- coding:utf-8 -*-
+from typing import Any, List, Mapping, Optional, Union
+
+import chromadb
+from chromadb.api import Collection
+from chromadb.config import Settings
+
+import os
+import sys
+sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+sys.path.append(os.path.dirname(os.path.abspath(__file__)))
+
+from config.config_parse import CHROMA_DB_PERSIST_PATH
+
+
+client = chromadb.Client(
+    Settings(
+        chroma_db_impl="duckdb+parquet",
+        persist_directory=CHROMA_DB_PERSIST_PATH,  # Optional, defaults to .chromadb/ in the current directory
+    )
+)

chat/python/instances/llm_instance.py

@@ -0,0 +1,21 @@
+# -*- coding:utf-8 -*-
+from langchain import llms
+
+import os
+import sys
+sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+sys.path.append(os.path.dirname(os.path.abspath(__file__)))
+
+from config.config_parse import LLM_PROVIDER_NAME, llm_config_dict
+
+
+def get_llm_provider(llm_provider_name: str, llm_config_dict: dict):
+    if llm_provider_name in llms.type_to_cls_dict:
+        llm_provider = llms.type_to_cls_dict[llm_provider_name]
+        llm = llm_provider(**llm_config_dict)
+        return llm
+    else:
+        raise Exception("llm_provider_name is not supported: {}".format(llm_provider_name))
+
+
+llm = get_llm_provider(LLM_PROVIDER_NAME, llm_config_dict)

chat/python/instances/logging_instance.py

@@ -0,0 +1,6 @@
+from loguru import logger
+import sys
+
+logger.remove() #remove the old handler. Else, the old one will work along with the new one you've added below'
+logger.add(sys.stdout, format="{time:YYYY-MM-DD at HH:mm:ss} | {level} | {message}", level="INFO")
+

chat/python/instances/text2vec.py

@@ -0,0 +1,23 @@
+# -*- coding:utf-8 -*-
+from typing import List
+
+from chromadb.api.types import Documents, EmbeddingFunction, Embeddings
+from langchain.embeddings import HuggingFaceEmbeddings
+
+from config.config_parse import HF_TEXT2VEC_MODEL_NAME
+
+hg_embedding = HuggingFaceEmbeddings(model_name=HF_TEXT2VEC_MODEL_NAME)
+
+
+class Text2VecEmbeddingFunction(EmbeddingFunction):
+    def __call__(self, texts: Documents) -> Embeddings:
+
+        embeddings = hg_embedding.embed_documents(texts)
+
+        return embeddings
+
+
+def get_embeddings(documents: List[str]) -> List[List[float]]:
+    embeddings = hg_embedding.embed_documents(documents)
+
+    return embeddings