metadata

title: Agents Course Final Assignment
emoji: 🕵🏻‍♂️
colorFrom: indigo
colorTo: indigo
sdk: gradio
sdk_version: 5.25.2
app_file: app.py
pinned: false
hf_oauth: true
hf_oauth_expiration_minutes: 480

Evaluation application for Unit 4 of the HuggingFace Agents course

This is my implementation of the evaluation application. Differently from the original application I cloned from, this one is thought to be run locally, because I am using Ollama.

When running locally, you have to create a .env file into the root of the project. This file gets read from the dotenv.load_dotenv() instruction and must contain the following variables:

HF_USERNAME="<your HuggingFace user name>"
HF_ACCESS_TOKEN="<your HuggingFace access token>"
SPACE_HOST="localhost"
SPACE_ID="<your space ID>"
TAVILY_API_KEY="<your tavily API key>"

You can infer the space ID by reading the address of your space when you access the Files section. For example, if you read

https://huggingface.co/spaces/aaa/bbb/tree/main

then the SPACE_ID is aaa/bbb (where aaa should be the user name).

Configuration

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference