open-predico dashboard

open-predico dashboard

Shows live forecasts and backtest results for the open-predico project as a static web page with charts. For people who follow its benchmark and want to see scores, target series and inputs.

Made by open-predico on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: This is a dashboard for the open-predico project. It shows live forecasts, a backtest leaderboard, the target series and a catalog of inputs. The page is static HTML that draws small data files, so it opens in a browser with nothing to install.

Who it suits: It suits people who follow the open-predico benchmark and want its forecasts and scores in one place.

Industry | Who uses it | How they use it | Example
Research | Forecasting researchers | Read backtest scores on the leaderboard | Look at how entries rank against each other
Data science | Analysts | Check the target series and input catalog | See which inputs feed a forecast
Open source | Project contributors | Review what the project publishes | Look at live forecasts after they are published
Teaching | Instructors | Show how forecasts get evaluated | Walk a class through a leaderboard

You can skip it if you want to build or run forecasting models yourself, since this page only displays results.

Where to use it:

• A web browser, on the Space's page at Hugging Face
• Your own computer, by opening the downloaded index.html in a browser

Use case | Who | What they get back
Follow live forecasts | Project followers | The latest forecasts, published after each gate
Compare backtests | Researchers | A leaderboard of backtest results
Inspect data | Analysts | The target series and the input catalog

How to set it up:

• Nothing to install if you use the Space online
• Python, to download the files with the Hugging Face tool
• A web browser to open index.html

1. Open the Space's page on Hugging Face to use it online.
2. To keep a copy, install huggingface_hub and download the Space into a folder.
3. Move into that folder.
4. Open index.html in a web browser.

pip install -U huggingface_hub
hf download open-predico/dashboard --repo-type space --local-dir dashboard
cd dashboard

How to use it: Open the page and look through its four views. The charts are drawn with uPlot from the small JSON files in the data folder. The README also shows one line that rebuilds the data in demo mode and serves the page locally.

Tool | What it does
Live forecasts | Shows forecasts published after each gate
Backtest leaderboard | Ranks results from backtests
Target series | Shows the series being forecast
Input catalog | Lists the inputs used

uv run python dashboard/build.py --demo-live && python -m http.server -d dashboard

Example: Open the dashboard and go to the backtest leaderboard. You see the backtest scores laid out so you can compare entries, and the live forecasts view shows what was published most recently.

For newcomers and experts: If you are new, open the Space online and click through the live forecasts first. Then try the leaderboard, which is where results from past tests sit.

If you know the field, the page is index.html plus app.js, drawing small JSON files with uPlot. A build script writes those files from the open-predico benchmark and live datasets on Hugging Face, with scores computed through the openpredico evaluation metrics. An upload option pushes the finished Space.

Build script reads the public datasets
It writes small JSON files to data
index.html and app.js load the files
uPlot draws the charts in the browser

Good to know:

• License: apache-2
• Cost to run: none named. The online page needs only a browser.
• Data it reaches: two public datasets on Hugging Face, a benchmark one and a live one, which the build step reads. The page draws small JSON files from its data folder.
• Not for: running or training forecasting models. It is a display page for results.

On Hugging Face, 13 November 2026

• Likes: 2
• Built with: static
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/open-predico/dashboard
• Files: https://huggingface.co/spaces/open-predico/dashboard/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.