This is an experimental app that uses the Huggingface Inference API with the gpt-neox-20b
model to generate Bash code from requests given in natural language. From time to time it will be improved trying to get better and better results :)
Inside app.py
replace <YOUR_ACCESS_TOKEN>
with your Huggingface's Access Token, then start the app with python3 app.py
To generate an access token register a new account at https://huggingface.co, confirm your email then, from your account go to Settings->Access Tokens and click on New token
and generate a new token with a rule " read". Use your token to try and test the app.
when you start the app you will be asked Print the Bash code needed to
... complete the sentence with the action to be performed and see the result