Use a binary classification model on Hugging Face to determine whether a statement in English is positive or negative.

goodbad.sh

#!/bin/bash
read -p "Hello! Give me something to work with! A sentence or tweet will do. " sentence

# now let's do some curl magic

curl --fail --silent --show-error -X POST -H "Authorization: Bearer HUGGINGFACE_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "'"$sentence"'"}' https://api-inference.huggingface.co/models/YOUR_USERNAME/PROJECT_URL | jq '.[0] | map(.) | if .[0].score < .[1].score then "positive" else "negative" end'