README.md

metadata

title: Store Sales Time Series Forecasting
emoji: 📈
colorFrom: blue
colorTo: green
sdk: docker
sdk_version: 1
app_file: app.py
pinned: false
short_description: Forecast store sales over 13 weeks using LSTM
license: mit
tags:
  - time-series
  - forecasting
  - lstm
  - retail
  - pytorch
library_name: pytorch
model-index:
  - name: ImprovedCashFlowLSTM
    results:
      - task:
          type: time-series-forecasting
        dataset:
          name: Kaggle Store Sales
          type: store-sales-time-series-forecasting
        metrics:
          - name: MAE
            type: mae
            value: 710.75
          - name: RMSE
            type: rmse
            value: 1108.51
          - name: MAPE
            type: mape
            value: 7.14
          - name: R2
            type: r2
            value: 0.8633

Store Sales Forecasting Model

Model Overview

This is an Improved Cash Flow LSTM model for forecasting store sales over a 13-week horizon, trained on the Kaggle Store Sales dataset.

Model Architecture

• Type: LSTM
• Input Size: 20
• Hidden Size: 128
• Number of Layers: 2
• Dropout: 0.2
• Forecast Horizon: 13 weeks
• Parameters: 227,085

Features Used

sales, onpromotion, trend, dayofweek_sin, dayofweek_cos, month_sin, month_cos, is_weekend, quarter, sales_lag_1, sales_lag_7, sales_lag_14, sales_ma_7, sales_ma_14, sales_ratio_7, sales_ratio_14, promo_lag_1, promo_running_sum, dcoilwtico, is_holiday

Performance Metrics

• MAE: $710.75
• RMSE: $1108.51
• MAPE: 7.14%
• R² Score: 0.8633

Usage

1. Install dependencies: pip install -r requirements.txt
2. Use inference.py to load the model and make predictions.
3. Input data must have 21 timesteps and match the feature set, scaled using scaler.pkl.

Example

import numpy as np
from inference import load_model_and_artifacts, forecast
model, scaler, feature_names = load_model_and_artifacts()
sequences = np.random.random((1, 21, len(feature_names)))  # Replace with actual data
predictions, uncertainties = forecast(model, scaler, sequences)
print("Predictions:", predictions)

Training Details

• Dataset: Kaggle Store Sales Time Series Forecasting
• Data Shape: (8250, 20)
• Sequences: 8217, length 21
• Preprocessing: StandardScaler, conservative outlier removal
• Epochs: 107 (early stopping)
• Optimizer: Adam with ReduceLROnPlateau
• Loss: MSE

Limitations

• Predictions are sensitive to input data quality.
• Model assumes consistent feature scaling.
• Performance may degrade on sparse or outlier-heavy data.

License

MIT License