feat: hugging face in demo

demo/.env_example

@@ -0,0 +1,2 @@
+HUGGINGFACEHUB_API_TOKEN=<your token>
+OPENAI_API_KEY=<your token>

examples/README.md → demo/README.md

@@ -4,6 +4,7 @@
 pip install -r requirements.txt
 ```
 
+Optional: Rename `.env.example` to `.env` and fill in the values.
 
 ### Running
 
@@ -13,4 +14,4 @@ streamlit run main.py
 
 ### Usage
 
-Access the application at http://localhost:8501
+Access the application at http://localhost:8501

demo/main.py

@@ -0,0 +1,64 @@
+from dotenv import load_dotenv
+import os
+import openai
+import streamlit as st
+
+# Load environment variables from .env file
+load_dotenv()
+
+# Get API keys from environment variables
+huggingface_api_key = os.getenv("HUGGINGFACEHUB_API_TOKEN")
+openai_api_key = os.getenv("OPENAI_API_KEY")
+
+with st.sidebar:
+    api_key = st.text_input(
+        "API Key", key="chatbot_api_key", type="password"
+    )
+    model_name = st.selectbox(
+        "Chat API Endpoint",
+        options=["gpt-4", "hf-gpt2"],
+        index=0,
+    )
+
+genoss_endpoint = "http://localhost:4321"
+
+st.title("🐂🌈 Genoss")
+if "messages" not in st.session_state:
+    st.session_state["messages"] = [
+        {"role": "assistant", "content": "How can I help you?"}
+    ]
+
+for msg in st.session_state.messages:
+    st.chat_message(msg["role"]).write(msg["content"])
+
+if prompt := st.chat_input():
+    st.session_state.messages.append({"role": "user", "content": prompt})
+    st.chat_message("user").write(prompt)
+    msg = ""
+
+    # Use the user-provided API key if available, otherwise use the API key from the .env file
+    if api_key == "" or api_key is None:
+        api_key = api_key if api_key else (huggingface_api_key if model_name.startswith("hf") else openai_api_key)
+        if api_key == "" or api_key is None:
+            st.error("Please provide an API key")
+            st.stop()
+
+    openai.api_key = api_key
+    openai.api_base = genoss_endpoint
+
+    try:
+        response = openai.ChatCompletion.create(
+            model=model_name,
+            messages=st.session_state.messages,
+        )
+        msg = response.choices[0].message
+    except Exception as e:
+        msg = f"Error: {e}"
+
+    st.empty()
+
+    st.session_state.messages.append(msg)
+    try:
+        st.chat_message("assistant").write(msg["content"])
+    except Exception as e:
+        st.error(f"Error: {e}, {msg}")

genoss/llm/base_genoss.py

@@ -1,4 +1,6 @@
 from abc import abstractmethod
+from ast import Not
+import re
 from typing import Dict, List
 
 from pydantic import BaseModel
@@ -10,11 +12,8 @@ class BaseGenossLLM(BaseModel):
 
     @abstractmethod
     def generate_answer(self, prompt: str) -> Dict:
-        print("You need to implement the generate_answer method")
-
-        return {id: 1}
+        return NotImplementedError
 
     @abstractmethod
     def generate_embedding(self, text: str) -> List[float]:
-        print("You need to implement the generate_embedding method")
-        return [0.0, 0.0, 0.0]
+        return NotImplementedError

genoss/llm/fake_llm.py

@@ -18,16 +18,12 @@ class FakeLLM(BaseGenossLLM):
     description: str = "Fake LLM for testing purpose"
 
     def generate_answer(self, question: str) -> Dict:
-        print("Generating Answer")
 
         llm = FakeListLLM(responses=["Hello from FakeLLM!"])
 
         llm_chain = LLMChain(llm=llm, prompt=prompt_template)
         response_text = llm_chain(question)
 
-        print("###################")
-        print(response_text)
-
         answer = response_text["text"]
         chat_completion = ChatCompletion(
             model=self.name, answer=answer, question=question

genoss/llm/hf_hub/base_hf_hub.py

@@ -32,8 +32,6 @@ def generate_answer(self, question: str) -> Dict:
         Generate answer from prompt
         """
 
-        print("Generating Answer...")
-
         llm = HuggingFaceHub(
             repo_id=self.repo_id, huggingfacehub_api_token=self.huggingfacehub_api_token
         )  # type: ignore

genoss/llm/local/gpt4all.py

@@ -17,16 +17,14 @@ class Gpt4AllLLM(BaseLocalLLM):
     model_path: str = "./local_models/ggml-gpt4all-j-v1.3-groovy.bin"
 
     def generate_answer(self, question: str) -> Dict:
-        print("Generating Answer")
 
         llm = GPT4All(
             model=self.model_path,  # pyright: ignore reportPrivateUsage=none
         )
 
         llm_chain = LLMChain(llm=llm, prompt=prompt_template)
         response_text = llm_chain(question)
-        print("###################")
-        print(response_text)
+
         answer = response_text["text"]
 
         chat_completion = ChatCompletion(

genoss/llm/openai/openai_llm.py

@@ -10,6 +10,7 @@
 from genoss.entities.chat.chat_completion import ChatCompletion
 from genoss.llm.base_genoss import BaseGenossLLM
 from genoss.prompts.prompt_template import prompt_template
+from langchain.chat_models import ChatOpenAI
 
 
 class OpenAILLM(BaseGenossLLM):
@@ -28,16 +29,12 @@ def __init__(self, model_name: str, api_key, *args, **kwargs):
         self.model_name = model_name
 
     def generate_answer(self, question: str) -> Dict:
-        print("Generating Answer")
 
-        llm = OpenAIChat(model_name=self.model_name, openai_api_key=self.openai_api_key)
+        llm = ChatOpenAI(model_name=self.model_name, openai_api_key=self.openai_api_key)
 
         llm_chain = LLMChain(llm=llm, prompt=prompt_template)
         response_text = llm_chain(question)
 
-        print("###################")
-        print(response_text)
-
         answer = response_text["text"]
         chat_completion = ChatCompletion(
             model=self.name, answer=answer, question=question