CodePeak 2025 : Supply Chain Demand Forecasting

[Mikeyzgoat]

Related Issue

• Demand Forecasting with LSTM-Transformer Model

• CodePeak 2025 Participant

• Contributor

Closes: #648 that will be closed through this PR

Describe the add-ons or changes you've made

Added a demand forecasting model using LSTM-Transformer to predict future sales quantities based on historical e-commerce data. Implemented data preprocessing, time series forecasting model, and visualization of results. Fine-tuned hyperparameters and applied a smoothing technique for better trend visualization.

Type of change

What sort of change have you made:

• New feature (non-breaking change which adds functionality)
• Bug fix (non-breaking change which fixes an issue)
• Code style update (formatting, local variables)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

How Has This Been Tested?

• Trained the LSTM-Transformer model for 500/1000 epochs with different learning rates.
• Validated results by comparing predicted vs. actual sales quantities.
• Visualized predictions using Matplotlib.
• Applied rolling forecasts for future predictions.

Checklist:

• My code follows the guidelines of this project.
• I have performed a self-review of my own code.
• I have commented my code, particularly wherever it was hard to understand.
• I have made corresponding changes to the documentation.
• My changes generate no new warnings.
• I have added things that prove my fix is effective or that my feature works.
• Any dependent changes have been merged and published in downstream modules.

Different Learning rates, epochs to check overfitting and underfitting

Training with epoch 500 and learning rate 0.01

Training with epoch 1000 and learning rate 0.0001

Training with epoch 1000 and learning rate 0.005

[welcome]

Hello there! 👋 Welcome to the project! 💖
Thank you and congrats 🎉 for opening your first pull request. Please adhere to our Code of Conduct. 🙌🏻 We will get back to you as soon as we can. 😄

Feel free to get in touch with me through social media handles. Hope to see you there!😄

[prathimacode-hub]

Hola, @Mikeyzgoat. Kindly use 3-4 algorithms and compare the models and accuracy. Also involve the lifecycle of data science : data analysis, data visualization, 3-4 models, compare models with accuracy and score followed by a summary statement mentioning best fit model out of implemented algorithms. If possible, add a .pkl file for best fit model. As of now, u had incorporated only LSTM model. Do the changes accordingly. Thank you.

[Mikeyzgoat]

Sure thing @prathimacode-hub , I will add SARIMAX and XGBOOST for comparision, prior to this I ll also run data cleaning, preprocessing and enhance feature extraction. Thanks for the review!, I will reach out within 1-2 days.

[prathimacode-hub]

Sure thing @prathimacode-hub , I will add SARIMAX and XGBOOST for comparision, prior to this I ll also run data cleaning, preprocessing and enhance feature extraction. Thanks for the review!, I will reach out within 1-2 days.

Perfect, go ahead..

[prathimacode-hub]

Any update? @Mikeyzgoat

[Mikeyzgoat]

Any update? @Mikeyzgoat

Yes, i might need a day or two, was caught up in some office work. Will revert back ASAP

[Mikeyzgoat]

Hi @prathimacode-hub sorry for the delay, i have added the models as asked and have followed the data preparation processes, kindly verify and update. Thanks!

[prathimacode-hub]

Your PR is approved. Well done.. 🙌 @Mikeyzgoat

[Mikeyzgoat]

Thank you @prathimacode-hub , can you update the same on CODEPEAK please, thanks again!. Happy coding

[prathimacode-hub]

Thank you @prathimacode-hub , can you update the same on CODEPEAK please, thanks again!. Happy coding

Sure, I asked for process to add tags, so shall do at earliest.. @Mikeyzgoat

[welcome]

Congrats on merging your first Pull Request! 🎉 All the best for your amazing open source journey ahead. 🚀⚡️