The Akash-Deployment-Gen CLI tool is designed to simplify the process of converting Docker Compose and Kubernetes deployments into Akash deployments. This tool automates the conversion, making it easier for developers to deploy their applications on the Akash decentralized cloud.
Before you begin, ensure you have the following installed:
- Python 3.6 or higher
- An active Akash wallet
- you need to create and fund an openai api account and generate an OPENAI_API_KEY ( we are still woring on the open source llm version of this )
Additionally, you must set the OPENAI_API_KEY and OPENAI_MODEL_NAME environment variables to use this CLI tool. You can obtain an API key from OpenAI.
To install the Auto-Akash-Deployments CLI tool, follow these steps:
- Create and activate a new Conda environment:
Ensure Conda is installed on your system. If not, follow the instructions on the official Conda installation guide.
After installing Conda, create a new environment by executing the following command in your terminal: conda create -n akash-deployment-gen && conda activate akash-deployment-gen
- Install the required Python packages:
After activating your Conda environment, install the necessary Python packages by running the following command in your terminal: pip install -r requirements.txt
This command will install all the dependencies listed in the requirements.txt file, ensuring that you have all the necessary libraries and tools to use the Auto-Akash-Deployments CLI tool.
- **Use cli tool to generate akash deployments **
Run an docker-compose example python akash_deployment_generation.py --input-file ./examples/dewy/dewy-docker-compose.yml --input-type docker-compose --github-repo-url https://github.com/DewyKB/dewy --output-file ./examples/dewy/akash_dewy.yml
Run an kubernetes example python deployment_generation.py --input-file ./examples/dewy/k8s.yml --input-type k8s --github-repo-url https://github.com/DewyKB/dewy --output-file ./examples/dewy/akash_dewy.yml
This tool doesn't currently support rebuilding docker images and their start scripts to use make some k8s or docker compose services work on akash.
If the k8s/docker compose container uses environment variables and ports in the entrypoint command this tool will not work without adding a new entrypoint script that can adjust the environment variables and port appropriately.
This code uses langgraph and the language agent tree search example to do the heavy lifting and duckduckgo for the search tool.
lats is the current best coding chain of thought approach that doesn't use a code interpreter.