UNSW-ZZSC9020 · laukamkit · Mar 9, 2026 · Mar 11, 2026 · Mar 11, 2026 · Mar 11, 2026
diff --git a/.DS_Store b/.DS_Store
diff --git a/.gitignore b/.gitignore
@@ -0,0 +1,11 @@
+*.venv
+#requirements*.txt
+
+src/checkpoint/*
+*.pyc
+__pycache__/
+
+*.pth
+*.npy
+
+timexer*
diff --git a/README.md b/README.md
@@ -1,4 +1,4 @@
-# Welcome to ZZSC9020 GitHub repository for group [GROUP-NAME]
+# Welcome to ZZSC9020 GitHub repository for group A
 
 This GitHub repository is the main point of access for students and lecturers of the ZZSC9020 capstone course. 
 
@@ -9,11 +9,11 @@ Complete the information below regarding your group.
 ## Group and project information
 
 ### Group members and zIDs
-- Member 1 (zID1) - Group leader
-- Member 2 (zID2) - role
-- Member 3 (zID3) - role
-- Member 4 (zID4) - role
-- Member 5 (zID5) - role
+- Brendan	(z5632728): Research consultant and Model analysis of SARIMAX + LSTM + Comparison
+- Daniel	(z5623647): Presentation/Communication Lead and Model analysis of Gradient Boosting + LSTM
+- Kam	(z5618444):     Project Manager, Presentation/Communication Lead and Model lead of PatchTST
+- Kelvin (z5502200):	Logistics coordinator, Code Refactorer, Model lead of PatchTST + SARIMAX + Comparison
+
 
 ### Brief project description
 
@@ -30,3 +30,25 @@ The repository has the following folder structure:
 - minutes: minutes for each meeting (left not more than 24h after the corresponding meeting)
 - report: RMarkdown or Jupyter notebook report in progress
 - src: source code
+    - ModelEvaluation: test loops on test data to retrieve RMSE and other metrics on best models
+    - ModelTuning: training/validation loops on best models to retrieve validation RMSE and other metrics on candidate models
+    - ModelFiles: code files that house different models we used to compare. 
+        - PatchTST_supervised: houses PatchTST code. Taken from PatchTST authors (yuqinie98, 2025) and adapted for our use case
+        - GradientBoosting / LSTM: additional utility functions for these models
+        - BasicModels.py / GroupAModels.py: holds the model classes. Gradient Booster and SARIMAX codes implemented by Scikit-learn.org., 2009 and ‌www.statsmodels.org respectively
+        - ModelConfigs.py: config classes used to standaradize inputs into the model classes
+        - ModelPlots.py: graphing functions to generate plots for the report
+    - NSWData: code files for datasets and dataloaders used commonly by the model classes
+        - NSWDataloader.py: extracts, pre-processes and train/val/test splitting of raw data files from data/NSW and saves it back into the folder
+        - NSWDataSet.py: torch data sets used in the training/testing loops of LSTM and PatchTST
+    - best_models.ipynb: notebook for analysis on validation data on candidate models to get best models for each family
+    - model_plots.ipynb: notebook for plotting predictions vs actuals used in the reporting
+    - model_test_results.ipynb: consolidates test metric results of best models and outputs tables for reporting
+    - others: draft notebooks used to refactor code and Exploratory Data Analysis notebook
+
+### Code References
+Scikit-learn.org. (2009). 3.2.4.3.6. sklearn.ensemble.GradientBoostingRegressor — scikit-learn 0.21.2 documentation. [online] Available at: https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingRegressor.html.
+
+‌www.statsmodels.org. (n.d.). statsmodels.tsa.statespace.sarimax.SARIMAX — statsmodels. [online] Available at: https://www.statsmodels.org/dev/generated/statsmodels.tsa.statespace.sarimax.SARIMAX.html.
+
+‌yuqinie98 (2025). GitHub - yuqinie98/PatchTST: An offical implementation of PatchTST: ‘A Time Series is Worth 64 Words: Long-term Forecasting with Transformers.’ (ICLR 2023) https://arxiv.org/abs/2211.14730. [online] GitHub. Available at: https://github.com/yuqinie98/PatchTST?tab=readme-ov-file [Accessed 14 Apr. 2026].
diff --git a/agendas/ZZSC9020 – Week 1 Project Meeting Agenda and Minutes (2nd Meeting).docx b/agendas/ZZSC9020 – Week 1 Project Meeting Agenda and Minutes (2nd Meeting).docx
diff --git a/agendas/ZZSC9020 – Week 1 Project Meeting Agenda.pdf b/agendas/ZZSC9020 – Week 1 Project Meeting Agenda.pdf
diff --git a/agendas/ZZSC9020 – Week 1B Project Meeting Agenda.docx b/agendas/ZZSC9020 – Week 1B Project Meeting Agenda.docx
diff --git a/agendas/ZZSC9020 – Week 2 Project Meeting Agenda and Minute.docx b/agendas/ZZSC9020 – Week 2 Project Meeting Agenda and Minute.docx
diff --git a/agendas/ZZSC9020 – Week 3 Project Meeting Agenda and Minute.docx b/agendas/ZZSC9020 – Week 3 Project Meeting Agenda and Minute.docx
diff --git a/agendas/ZZSC9020 – Week 4 Project Meeting Agenda and Minute.docx b/agendas/ZZSC9020 – Week 4 Project Meeting Agenda and Minute.docx
diff --git a/agendas/ZZSC9020 – Week 5 Project Meeting Agenda and Minute.docx b/agendas/ZZSC9020 – Week 5 Project Meeting Agenda and Minute.docx
diff --git a/data/NSW/df_model.csv b/data/NSW/df_model.csv
diff --git a/data/NSW/scaler.pkl b/data/NSW/scaler.pkl
diff --git a/data/NSW/test.csv b/data/NSW/test.csv
diff --git a/data/NSW/test_scaled.csv b/data/NSW/test_scaled.csv
diff --git a/data/NSW/train.csv b/data/NSW/train.csv