feat(examples): update setting to run demo (#968)

* update to run demo

* update retriever

* update

* fix pre-commit

* update readme

.env.template

@@ -1,5 +1,5 @@
 # openai keys
-OPENAI_API_KEY="xxx"
+OPENAI_API_KEY="sk-xxxxx"
 
 # local storage
-DATABASE_FOLDER_PATH="xxx"
+DATABASE_FOLDER_PATH="./data"

README-CN.md

@@ -32,7 +32,7 @@
 通过 `pypi` 安装 `research-town` ：
 
 ```bash
-pip install research-town 
+pip install research-town
 ```
 
 ### 虚拟环境安装
@@ -50,7 +50,7 @@ curl -sSL https://install.python-poetry.org | python3
 1. 将 `.env.template` 文件复制到项目根目录，并命名为 `.env`。
 ```bash
 cp .env.template .env
-``` 
+```
 2. 在 `.env` 文件中填写所需的环境变量。
 
 ### 运行示例
@@ -114,4 +114,3 @@ pre-commit install
   </picture>
 </a>
 </p>
-

README.md

@@ -47,6 +47,7 @@ Use a virtual environment, e.g. with anaconda3:
 conda create -n research-town python=3.10
 conda activate research-town
 curl -sSL https://install.python-poetry.org | python3
+export PATH="$HOME/.local/bin:$PATH"
 ```
 
 ### Configure environment variables
@@ -66,6 +67,8 @@ cd examples
 python research_town_demo.py
 ```
 
+All generated research progress like ideas, proposals will be automatically saved in the database folder you set in the .env file.
+
 ## Developing
 
 #### Develop Demo

pyproject.toml

@@ -22,9 +22,9 @@ lxml = "*"
 tqdm = "*"
 keybert = "*"
 semanticscholar = "*"
-torch = { version = ">=2.2.0, <2.3.0", optional = true }
-
+socksio = "*"
 
+torch = { version = ">=2.2.0, <2.3.0", optional = true }
 uvicorn = { version = "*", optional = true }
 fastapi = { version = "*", optional = true }
 pandas = { version = "*", optional = true }

research_town/dbs/db_base.py

@@ -37,9 +37,7 @@ def add(self, data: T) -> None:
         self.database_client.add(
             self.data_class.__name__, [data.model_dump(exclude_none=True)], embeddings
         )
-        logger.info(
-            f"Creating instance of '{data.__class__.__name__}': '{data.model_dump()}'"
-        )
+        logger.info(f"Creating instance of '{data.__class__.__name__}'")
 
     def update(self, pk: str, updates: Dict[str, Any]) -> bool:
         return self.database_client.update(self.data_class.__name__, pk, updates)

research_town/utils/retriever.py

@@ -6,10 +6,8 @@
 
 def get_embed(
     instructions: List[str],
-    retriever_tokenizer: BertTokenizer = BertTokenizer.from_pretrained(
-        'facebook/contriever'
-    ),
-    retriever_model: BertModel = BertModel.from_pretrained('facebook/contriever'),
+    retriever_tokenizer: BertTokenizer,
+    retriever_model: BertModel,
 ) -> List[torch.Tensor]:
     encoded_input_all = [
         retriever_tokenizer(text, return_tensors='pt', truncation=True, max_length=512)