Sales Prediction with time series analysis and forecasting

Analyse and predict the sales of a toy manufacturing company using sophisticated data science methods (time-series analysis and forecasting).
Explore the docs »

View Demo · Report Bug · Request Feature

Table of Contents

• About the Project
  • Built With
• Getting Started
  • Data Exploration
  • Extract Transform Load
  • Model definition, training, and evaluation
• Usage
• Roadmap
• Contributing
• License
• Contact
• Acknowledgements

About The Project

This is a Data Science project using C1 Company sales data.
This project consists of data analysis, model definition and model training and evaluation using various data science packages.

Built With

• NumPy
• Matplotlib
• Seaborn
• Statmodels
• Keras
• Tensorflow
• Sci-Kit Learn
• Pandas
• LightGBM
• XGBoost

Getting Started

The project consists of three stages:

1. Data Exploration
2. Extract Transform Load
3. Model definition, training, and evaluation

The .ipynb (Jupyter Notebook) files contain the three stages of the project.

Data exploration

Data Exploration refers to the initial exploration of the dataset along with subtle data analysis to obtain a vague idea about the dataset.
This consists of statistical description, correlation matrix, box plots, etc.

Extract Transform Load

Extract Transform Load (ETL) refers to the process of copying data from one or more sources into a destination system which represents the data differently from the source or in a different context than the source.
It also involves transforming the data, feature engineering, detailed data analysis and improving the overall quality of the data.

Model definition, training, and evaluation

Model definition refers to the process of choosing the models those are best suited for the dataset, along with the initial hyperparameters.
Model training refers to the process of fitting the models to the training data.
Model evaluation refers to the process of evaluating the process of the models using certain appropriate evaluation metrics and tuning the hyperparameters again based on the results.

It is an iterative process until required performance is obtained.

Usage

This project analyses time-series data and predicts future data using machine learning models trained over the same.
This project approach can be used for the following:

• Weather forecasting
• Stock price prediction
• Sales analysis
• Business reformation

Roadmap

See the open issues for a list of proposed features (and known issues).

Contributing

Contributions are what make the open source community such an amazing place to be learn, inspire, and create. Any contributions you make are greatly appreciated.

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

Distributed under the MIT License. See LICENSE for more information.

Contact

IE-Mechatronics - @iemechatronics
E-Mail - convener.iemechatronics@gmail.com

Project Link: https://github.com/iemct/Sales_Prediction

Acknowledgements

• 1C Company for providing the dataset for public use.
• Vamsi Vasamsetti for building the ETL notebook.
• Supriti Vijay for building the model implementation notebook.
• Rikin Ramachandran for working on the ETL notebook.
• Soundar Murugan for creating the architectural decisions document.