Quick Start Guide - OpenCodeReview

Deployment Instructions

This guide walks you through the process of setting up and deploying the OpenCodeReview platform on your local environment. Follow these steps carefully to ensure a successful installation.

Requirements

Docker and Docker Compose installed on your system
Minimum 8GB RAM and 4 CPU cores recommended for optimal performance
Preferably a Linux-based OS (Ubuntu, CentOS, etc.). Windows and macOS are supported but may require additional configuration
API keys configured in .env file. OpenAI and NVIDIA API keys are required for the application to function properly
Docker image nvcr.io/nvidia/base/ubuntu - you should have login access to this image

Installation Steps

Create Installation Directory
```
mkdir -p /hackathon/
```
Download and Extract Files
Choose one of the following options:
1. Full Package Download (Recommended)
  Download the full package file:
  
  Download Full Package (~630MB)
2. Package Extraction
  Save the downloaded application tgz file to /hackathon/ and extract it:
```
tar -xvzf <filename>.tgz
```
  After extraction, a folder named opencode-review will be created.
Login to NVIDIA Docker Registry
Before starting the containers, you must log in to the NVIDIA container registry to access the required Docker image.
See the official instructions here: NVIDIA Docker Registry Login Guide
```
$ docker login nvcr.io
Username: $oauthtoken
Password: < Insert NVIDIA_API_KEY here >
```
You will need your NVIDIA NGC credentials. If you do not have an account, follow the instructions in the link above.
Configure API Keys
Edit the .env file to set your API keys:
```
# Example .env file
NVIDIA_API_KEY=your_nvidia_api_key_here
OPENAI_API_KEY=your_openai_api_key_here
```
Note: Both NVIDIA_API_KEY and OPENAI_API_KEY are required for the application to function properly.
Start the Docker Containers
```
./restart.sh
```
This script will prompt for Docker login credentials if necessary to access the nvcr.io/nvidia/base/ubuntu image.
Access the Application
Once deployment is complete, access the OpenCodeReview platform at:

http://localhost:5174

Troubleshooting

Common Issues

Docker Login Failure:
Ensure you have valid credentials for accessing the NVIDIA container registry.
Resource Constraints:
If the application runs slowly, ensure your system meets the recommended specifications (8GB RAM, 4 CPU cores).
API Key Configuration:
Verify that your API keys in the .env file are correctly formatted and valid.

Getting Help

If you continue to experience issues with the installation, please:

Check the GitHub Issues page for existing solutions
Create a new issue if your problem hasn't been addressed
Reach out via the contact information provided in the GitHub repository

Ready to get started? Watch our quick demo video to see OpenCodeReview in action:

View Project on GitHub