Upload DATASET_DESC.md with huggingface_hub

DATASET_DESC.md

@@ -2,4 +2,62 @@
 The public scenes, scenarios, and vehicles can be downloaded from [this link](https://huggingface.co/datasets/XDimLab/HUGSIM). You are free to use any data that is helpful for training the model, including whether or not to use the training set we provide.
 
 ## Testing Set
-To algorithms avoid overfitting to the testing scenarios, we do not provide any data from the testing set used for evaluation in this competition.
+To algorithms avoid overfitting to the testing scenarios, we do not provide any data from the testing set used for evaluation in this competition.
+
+## Locally Testing
+
+This guide walks you through how to run the husim_server locally. If you can complete this setup on your machine, the same approach will work in the competition environment.
+
+**1. Clone the Repository**
+
+Clone server code:
+
+```bash
+git clone --branch hf-competitions --single-branch [email protected]:longxiaofei/HUGSIM.git
+```
+
+**2. Build the Docker Image**
+
+Navigate to the project directory and build the Docker image:
+
+```bash
+docker build . -f ./docker/web_server_dockerfile -t hugsim_server:local
+```
+
+**3. Run the Docker Container**
+
+Run the server locally on port 7860:
+
+```bash
+docker run -d -p 7860:7860 --name hugsim_server hugsim_server:local
+```
+
+
+**4. Install the Client**
+
+Install the hugsim_client package:
+
+```bash
+pip install --upgrade hugsim_client
+```
+
+
+**5. Using the Client**
+
+```python
+from hugsim_client import HugsimClient
+
+client = HugsimClient()
+
+# Get the current state of the environment
+client.get_current_state()
+
+# Reset the environment
+client.reset_env()
+
+# Execute an action in the environment (replace `xxx` with your action)
+client.execute_action(xxx)
+```
+
+The full example: [ltf_e2e.py](https://huggingface.co/Libra-1995/navsim/blob/main/ltf_e2e.py)
+