Final Submission: UC00213 Urban Pedestrian Climate Impact Prediction

[Thivainv]

Summary
This pull request represents the final submission for the deep learning
component of the UC00213 Urban Pedestrian Climate Impact Prediction project.
Three files have been added to the repository as required for the ready to
publish submission.

Files Added

• UC00213_Urban_Pedestrian_Climate_Impact_Prediction.ipynb The completed
  project notebook including Section 10 (Deep Learning) with all code cells,
  markdown documentation, inline citations, and references fn-53 to fn-63.
• UC00213_Urban_Pedestrian_Climate_Impact_Prediction.html HTML export of
  the full notebook for browser-based viewing without requiring Jupyter.
• UC00213_Urban_Pedestrian_Climate_Impact_Prediction.json Project metadata
  file describing the use case, technologies, datasets, and tags.

Deep Learning Section Summary

• Three recurrent neural network architectures implemented and compared:
  • Baseline LSTM (24,385 parameters)
  • Stacked LSTM (35,777 parameters)
  • GRU (19,009 parameters)
• Best model: GRU with R² of 0.9407, MAE of 4,921.98, RMSE of 7,325.04
• Full markdown documentation with citations throughout
• Reproducibility ensured via fixed random seed and model save/load
• Notebook runs clean end to end with Kernel , Restart & Run All

Checklist

• Notebook runs without errors from top to bottom
• All markdown cells include inline citations
• References fn-53 to fn-63 added to reference list
• Models saved and loaded for reproducible evaluation
• HTML export generated and verified
• JSON metadata file included

[molliefernandez-mentor]

Hi Thivain, just a few changes before you can merge:

• Please delete the 'test' file
• Should there be 2 authors listed didn't you work on this together?
• there's an error in code block 109
• Can you please bold the title parts of the intro like 'scenario' and 'introduction' etc.

[Thivainv]

Hi Mollie, I have made all the requested changes:

• Fixed the input_shape warning in code block 109 using Input layer
• Bolded all intro section titles
• Added both authors to the notebook
• Deleted the test file
• Added HTML and JSON files

Ready for review.

[molliefernandez-mentor]

Thanks Thivain looks great!