Pull Request: Integrate .env File Support and Add Backend Usage Examp…

…le (InternLM#187)

* feat: Integrate .env file support for environment variables
feat: Add backend_example.py file support for using without a frontend

- Added .env.example with necessary environment variable placeholders: OPENAI_API_KEY, OPENAI_API_BASE, OPENAI_MODEL, SILICON_API_KEY, SILICON_MODEL.
- Updated models.py to load environment variables using python-dotenv.
- Modified gpt4 amd silicon model configurations to use values from .env file or defaults.
- Updated .gitignore to exclude .env file.
- Updated requirements.txt to include python-dotenv.
- Updated README.md to document environment variable setup and backend usage example.
- Added backend_example.py for direct backend interaction.

* (README.md): remove new features section from forked version

.env.example

@@ -0,0 +1,5 @@
+OPENAI_API_KEY=
+OPENAI_API_BASE=
+OPENAI_MODEL=
+SILICON_API_KEY=
+SILICON_MODEL=

.gitignore

@@ -160,4 +160,6 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
+
+.env
 temp

README.md

@@ -15,29 +15,6 @@ English | [简体中文](README_zh-CN.md)
 
 ## ✨ MindSearch: Mimicking Human Minds Elicits Deep AI Searcher
 
-MindSearch is an open-source AI Search Engine Framework with Perplexity.ai Pro performance. You can simply deploy it with your own perplexity.ai style search engine with either close-source LLMs (GPT, Claude) or open-source LLMs ([InternLM2.5 series](https://huggingface.co/internlm/internlm2_5-7b-chat) are specifically optimized to provide superior performance within the MindSearch framework; other open-source models have not been specifically tested). It owns following features:
-
-- 🤔 **Ask everything you want to know**: MindSearch is designed to solve any question in your life and use web knowledge.
-- 📚 **In-depth Knowledge Discovery**: MindSearch browses hundreds of web pages to answer your question, providing deeper and wider knowledge base answer.
-- 🔍 **Detailed Solution Path**: MindSearch exposes all details, allowing users to check everything they want. This greatly improves the credibility of its final response as well as usability.
-- 💻 **Optimized UI Experience**: Providing all kinds of interfaces for users, including React, Gradio, Streamlit and Terminal. Choose any type based on your need.
-- 🧠 **Dynamic Graph Construction Process**: MindSearch decomposes the user query into atomic sub-questions as nodes in the graph and progressively extends the graph based on the search result from WebSearcher.
-
-<div align="center">
-
-<img src="assets/teaser.gif">
-
-</div>
-
-## ⚡️ MindSearch vs other AI Search Engines
-
-Comparison on human preference based on depth, breadth, factuality of the response generated by ChatGPT-Web, Perplexity.ai (Pro), and MindSearch. Results are obtained on 100 human-crafted real-world questions and evaluated by 5 human experts\*.
-
-<div align="center">
-<img src="assets/mindsearch_openset.png" width="90%">
-</div>
-* All experiments are done before July.7 2024.
-
 ## ⚽️ Build Your Own MindSearch
 
 ### Step1: Dependencies Installation
@@ -48,7 +25,16 @@ cd MindSearch
 pip install -r requirements.txt
 ```
 
-### Step2: Setup MindSearch API
+### Step2: Setup Environment Variables
+
+Before setting up the API, you need to configure environment variables. Rename the `.env.example` file to `.env` and fill in the required values.
+
+```bash
+mv .env.example .env
+# Open .env and add your keys and model configurations
+```
+
+### Step3: Setup MindSearch API
 
 Setup FastAPI Server.
 
@@ -69,7 +55,7 @@ python -m mindsearch.app --lang en --model_format internlm_server --search_engin
 
   Please set your Web Search engine API key as the `WEB_SEARCH_API_KEY` environment variable unless you are using `DuckDuckGo`.
 
-### Step3: Setup MindSearch Frontend
+### Step4: Setup MindSearch Frontend
 
 Providing following frontend interfaces,
 
@@ -104,6 +90,35 @@ python frontend/mindsearch_gradio.py
 streamlit run frontend/mindsearch_streamlit.py
 ```
 
+## 🌐 Change Web Search API
+
+To use a different type of web search API, modify the `searcher_type` attribute in the `searcher_cfg` located in `mindsearch/agent/__init__.py`. Currently supported web search APIs include:
+
+- `GoogleSearch`
+- `DuckDuckGoSearch`
+- `BraveSearch`
+- `BingSearch`
+
+For example, to change to the Brave Search API, you would configure it as follows:
+
+```python
+BingBrowser(
+    searcher_type='BraveSearch',
+    topk=2,
+    api_key=os.environ.get('BRAVE_API_KEY', 'YOUR BRAVE API')
+)
+```
+
+## 🐞 Using the Backend Without Frontend
+
+For users who prefer to interact with the backend directly, use the `backend_example.py` script. This script demonstrates how to send a query to the backend and process the response.
+
+```bash
+python backend_example.py
+```
+
+Make sure you have set up the environment variables and the backend is running before executing the script.
+
 ## 🐞 Debug Locally
 
 ```bash

backend_example.py

@@ -0,0 +1,34 @@
+import json
+import requests
+
+# Define the backend URL
+url = 'http://localhost:8002/solve'
+headers = {'Content-Type': 'application/json'}
+
+# Function to send a query to the backend and get the response
+def get_response(query):
+    # Prepare the input data
+    data = {'inputs': [{'role': 'user', 'content': query}]}
+
+    # Send the request to the backend
+    response = requests.post(url, headers=headers, data=json.dumps(data), timeout=20, stream=True)
+
+    # Process the streaming response
+    for chunk in response.iter_lines(chunk_size=8192, decode_unicode=False, delimiter=b'\n'):
+        if chunk:
+            decoded = chunk.decode('utf-8')
+            if decoded == '\r':
+                continue
+            if decoded[:6] == 'data: ':
+                decoded = decoded[6:]
+            elif decoded.startswith(': ping - '):
+                continue
+            response_data = json.loads(decoded)
+            agent_return = response_data['response']
+            node_name = response_data['current_node']
+            print(f"Node: {node_name}, Response: {agent_return['response']}")
+
+# Example usage
+if __name__ == '__main__':
+    query = "What is the weather like today in New York?"
+    get_response(query)

mindsearch/agent/models.py

@@ -1,8 +1,11 @@
 import os
+from dotenv import load_dotenv
 
 from lagent.llms import (GPTAPI, INTERNLM2_META, HFTransformerCasualLM,
                          LMDeployClient, LMDeployServer)
 
+load_dotenv()
+
 internlm_server = dict(type=LMDeployServer,
                        path='internlm/internlm2_5-7b-chat',
                        model_name='internlm2',
@@ -36,7 +39,7 @@
                    stop_words=['<|im_end|>'])
 # openai_api_base needs to fill in the complete chat api address, such as: https://api.openai.com/v1/chat/completions
 gpt4 = dict(type=GPTAPI,
-            model_type='gpt-4-turbo',
+            model_type=os.environ.get('OPENAI_MODEL', 'gpt-4o'),
             key=os.environ.get('OPENAI_API_KEY', 'YOUR OPENAI API KEY'),
             openai_api_base=os.environ.get('OPENAI_API_BASE', 'https://api.openai.com/v1/chat/completions'),
             )
@@ -60,7 +63,7 @@
             stop_words=['<|im_end|>'])
 
 internlm_silicon = dict(type=GPTAPI,
-                        model_type='internlm/internlm2_5-7b-chat',
+                        model_type=os.environ.get('SILICON_MODEL', 'internlm/internlm2_5-7b-chat'),
                         key=os.environ.get('SILICON_API_KEY', 'YOUR SILICON API KEY'),
                         openai_api_base='https://api.siliconflow.cn/v1/chat/completions',
                         meta_template=[

requirements.txt

@@ -10,3 +10,4 @@ sse-starlette
 termcolor
 transformers==4.41.0
 uvicorn
+python-dotenv