README.md

metadata

title: DATFID MASTER
emoji: 🤖
colorFrom: indigo
colorTo: pink
sdk: docker
pinned: false
short_description: DATFID - Secure and powerful forecasting SDK

DATFID MASTER DEMO (Hackathon Proxy)

datfid-master is the public proxy layer. For the hackathon setup, it forwards all model calls to datfid_api and does not require the webpage demo flow.

We achieved state-of-the-art accuracy in the M5 Forecasting Competition, beated all the benchmarks, forecasting hierarchical Walmart product sales over a 5-year period using advanced exogenous features and rigorous error metrics (Kaggle M5 page).

---

Hackathon Setup

• Set hf_url to your public datfid_api-demo Space URL.
• No user/API token is required in this demo setup.
• Students call only datfid-master; orchestration stays unchanged.
• Backend internals in datfid_api-demo are simplified for demo use.

---

Getting Started

1. Install the SDK

pip install -i https://test.pypi.org/simple/ datfid

2. Example Usage

import pandas as pd
from datfid.client import DATFIDClient

# Initialize client (no token needed in demo)
client = DATFIDClient()

# Fit the model
fit_result = client.fit_model(
    df=my_dataframe,
    id_col="SKU_ID",
    time_col="Date",
    y="Sales",
    lagged_features={"Sales": 2, "Promo": 1},
    current_features=["Price", "Holiday"],
    filter_by_significance=True,
    meanvar_test=True
)

# Forecast
forecast_df = client.forecast_model(df_forecast=my_forecast_dataframe)
print(forecast_df.head())

# Check service health
client.secure_ping()

---

Contact & Info

• Website: datfid.com
• Demo setup: no token required

---

Happy forecasting with confidence!
— The DATFID Team