Welcome to our repository for the Bakery Shop Sales Prediction Challenge, a practice competition for Predicta 1.0 organized by the IEEE Student Branch, University of Peradeniya. This competition aims to test and improve our forecasting skills by predicting daily sales quantities of various bakery items across three different outlets.
In the food industry, accurately predicting product demand is crucial for maintaining efficiency and ensuring freshness. Our task is to forecast the daily sales quantities of bakery items from June 1st to June 14th, 2024, using historical sales data from January 1st to May 31st, 2024.
-
historical_data.csv: Contains historical sales data with the following columns:
transaction_id
: Unique identifier for each transactiontransaction_date
: Date of the transactiontransaction_time
: Time of the transactiontransaction_qty
: Quantity of items sold in the transactionstore_id
: Identifier for the store where the transaction occurredproduct_id
: Identifier for the product soldunit_price
: Price per unit of the product
-
product_descriptions.csv: Contains descriptions of the products with the following columns:
product_id
: Identifier for the productproduct_category
: Category of the product (e.g., Pastries, Cakes)product_type
: Type of the product (e.g., Croissants, Muffins)product_detail
: Detailed description of the product
-
submission_key.csv: Template for our submission with the following columns:
ID
: ID assigned for the respective date, shop, and product combinationtransaction_date
: Date for which the sales quantity is predictedstore_id
: Identifier for the storeproduct_id
: Identifier for the product
-
submission_format.csv: Submission format with the following columns:
ID
: ID assigned for the respective date, shop, and product combinationsold_qty
: Quantity of items sold (to be filled by participants)
Submissions are evaluated based on the Root Mean Squared Error (RMSE) between the predicted sold quantities and the actual sold quantities.
- Member 1: Akhila Prabodha
- Member 2: Navini Jagoda
- Member 3: Janindu Shehan
- Python 3.x
- Jupyter Notebook
- Required Python libraries (see
requirements.txt
)
Clone the repository and install the required packages:
git clone https://github.com/yourusername/bakery-sales-prediction.git
cd bakery-sales-prediction
pip install -r requirements.txt
- Data Analysis: Explore the data using
notebooks/data_analysis.ipynb
. - Feature Engineering: Create new features in
notebooks/feature_engineering.ipynb
. - Model Training: Train your predictive models in
notebooks/model_training.ipynb
. - Model Evaluation: Evaluate model performance in
notebooks/model_evaluation.ipynb
. - Prediction: Generate predictions using
src/predict.py
and create the final submission file.
Submit the final_submission.csv
file located in the submissions/
directory.
We would like to thank the IEEE Student Branch, University of Peradeniya for organizing this competition and providing us with the opportunity to enhance our forecasting skills.