You can deploy this model and start earning money today!

JunLim99/
dummy-hw3-part2

Name: dummy-hw3-part2
Availability: InStock

Views

No views yet

DummyModel for HW3 Part 2

This is a tiny dummy "model" repo designed to make it easy to score the HW3 Part 2 meta‑metrics: ramp‑up time, bus factor proxy, performance claim, license, dataset & code score, dataset quality, and code quality.

Quickstart

bash

python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
pytest -q
python examples/quickstart.py

Usage

python

from dummy_model.model import DummyModel
m = DummyModel()
print(m.predict("hello world"))

Performance Claim (toy)

On a synthetic 10‑example dataset (see data/toy.csv), DummyModel achieves Accuracy = 1.0 by always predicting the label "positive". This is purely for demonstration and does not reflect real performance.

Dataset & Code

Code lives in dummy_model/ with a simple DummyModel class.
A toy dataset with schema (text:str, label:str) is provided at data/toy.csv.
Example script in examples/quickstart.py.
Unit tests in tests/ and CI in .github/workflows/ci.yml.

Dataset Quality (toy)

Schema: columns = text, label (string).
Splits: for demonstration, treat first 8 rows as "train" and last 2 rows as "test".
Provenance: generated locally for HW3; no external data was used.

License

MIT (see LICENSE). On Hugging Face, set the repo License field to mit.

Notes

This repo is intentionally minimal so it can be published to Hugging Face as a model repo.

JunLim99/dummy-hw3-part2