Merge pull request #783 from valentimarco/develop

[HUGE PR] Chat Messages for ChatModels

core/cat/convo/messages.py

@@ -1,29 +1,61 @@
-
-
-
-
-
 from typing import List, Dict
 from cat.utils import BaseModelDict
+from langchain_core.messages import BaseMessage, AIMessage, HumanMessage
+from enum import Enum
 
-#class WorkingMemory(BaseModelDict):
-#    history : List = []
 
+class Role(Enum):
+    AI = "AI"
+    Human = "Human"
 
 class MessageWhy(BaseModelDict):
+    """Class for wrapping message why
+
+    Variables:
+        input (str): input message
+        intermediate_steps (List): intermediate steps
+        memory (dict): memory
+    """
     input: str
     intermediate_steps: List
     memory: dict
 
 
 class CatMessage(BaseModelDict):
+    """Class for wrapping cat message
+
+    Variables:
+        content (str): cat message
+        user_id (str): user id
+    """
     content: str
     user_id: str
     type: str = "chat"
     why: MessageWhy | None = None
 
 
 class UserMessage(BaseModelDict):
+    """Class for wrapping user message
+
+    Variables:
+        text (str): user message
+        user_id (str): user id
+    """
     text: str
     user_id: str
 
+
+
+def convert_to_Langchain_message(messages: List[UserMessage | CatMessage] ) -> List[BaseMessage]:
+    messages = []
+    for m in messages:
+        if isinstance(m, CatMessage):
+            messages.append(HumanMessage(content=m.content, response_metadata={"userId": m.user_id}))
+        else:
+            messages.append(AIMessage(content=m.text, response_metadata={"userId": m.user_id}))
+    return messages
+
+def convert_to_Cat_message(cat_message: AIMessage, why: MessageWhy) -> CatMessage:
+    return CatMessage(content=cat_message.content, user_id=cat_message.response_metadata["userId"], why=why)
+
+

core/cat/factory/custom_llm.py

@@ -8,8 +8,9 @@
 
 from langchain_core.language_models.llms import LLM
 from langchain_openai.llms import OpenAI
-from langchain_community.llms.ollama import Ollama, OllamaEndpointNotFoundError
+from langchain_community.llms.ollama import Ollama,OllamaEndpointNotFoundError
 from langchain_openai.chat_models import ChatOpenAI
+from langchain_community.chat_models.ollama import ChatOllama
 
 from cat.log import log
 
@@ -91,7 +92,7 @@ def __init__(self, **kwargs):
 
 
 
-class CustomOllama(Ollama):
+class CustomOllama(ChatOllama):
     def __init__(self, **kwargs: Any) -> None:
         if "localhost" in kwargs["base_url"]:
             log.error(

core/cat/factory/llm.py

@@ -8,6 +8,7 @@
 )
 from langchain_openai import ChatOpenAI
 from langchain_google_genai import ChatGoogleGenerativeAI
+from langchain_community.chat_models.ollama import ChatOllama
 
 from .ollama_utils import _create_stream_patch, _acreate_stream_patch
 from typing import Type
@@ -18,6 +19,7 @@
 from cat.mad_hatter.mad_hatter import MadHatter
 
 
+
 # Base class to manage LLM configuration.
 class LLMSettings(BaseModel):
     # class instantiating the model
@@ -228,29 +230,21 @@ class LLMHuggingFaceEndpointConfig(LLMSettings):
             "link": "https://huggingface.co/inference-endpoints",
         }
     )
-
-
-# monkey patch to fix stops sequences
-OllamaFix: Type = CustomOllama
-OllamaFix._create_stream = _create_stream_patch
-OllamaFix._acreate_stream = _acreate_stream_patch
-
-
 class LLMOllamaConfig(LLMSettings):
     base_url: str
-    model: str = "llama2"
+    model: str = "llama3"
     num_ctx: int = 2048
     repeat_last_n: int = 64
     repeat_penalty: float = 1.1
     temperature: float = 0.8
 
-    _pyclass: Type = OllamaFix
+    _pyclass: Type = CustomOllama
 
     model_config = ConfigDict(
         json_schema_extra={
             "humanReadableName": "Ollama",
             "description": "Configuration for Ollama",
-            "link": "https://ollama.ai/library",
+            "link": "https://ollama.ai/library"
         }
     )