Add T1 2026 Orion project structure and resolve pandas/frontend dependency issues

26_T1/afl_player_tracking_and_crowd_monitoring/BoundingBoxes

@@ -0,0 +1,89 @@
+# -*- coding: utf-8 -*-
+"""boundingboxeswithevents.ipynb
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1qjF8riAPbImWdy3NhnVNWT4C_vacw1j-
+"""
+
+import cv2
+import json
+import numpy as np
+import pandas as pd
+
+kick = pd.read_csv('kick_synced_annotations.csv')
+kick = pd.DataFrame(kick)
+
+kick.head()
+
+#reading the event data and storing it as a pandas data frame
+event = pd.read_csv('mark_synced_annotations.csv')
+event = pd.DataFrame(event)
+
+
+#reading the tracking data
+with open('mark_1 1track.json', 'r') as f:
+    annotations = json.load(f)
+
+# reading the raw video
+video_path = 'mark_1.mp4'
+cap = cv2.VideoCapture(video_path)
+
+
+fps = int(cap.get(cv2.CAP_PROP_FPS))
+width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
+height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
+
+# output video
+output_path = 'mark_1_event.mp4'
+fourcc = cv2.VideoWriter_fourcc(*'mp4v')
+out = cv2.VideoWriter(output_path, fourcc, fps, (width, height))
+
+
+# initiating the frame index as 0
+frame_index = 0
+
+
+
+while cap.isOpened():
+    ret, frame = cap.read()
+    if not ret:
+        break
+    # iterating through traking data for each frame
+    if annotations['tracking_results'][frame_index]['frame_number'] == frame_index + 1:
+      #iterating through each object in the tracking results
+        for box in annotations['tracking_results'][frame_index]['players']:
+            x1, y1, x2, y2 = box['bbox']['x1'], box['bbox']['y1'], box['bbox']['x2'], box['bbox']['y2']
+            # Rectangle around the player, co-ordinants of box (x1, y1), (x2, y2), Box colour (255, 255, 0) Box Thickness = 2
+            cv2.rectangle(frame, (x1, y1), (x2, y2), (255, 255, 0), 2)
+            # box for label below the player
+            cv2.rectangle(frame, (x1, y2), (x1 +115, y2 +50), (255, 255, 0), -1)
+            # label for player only displaying the player ID to prevent the frame being over populated with information
+            cv2.putText(frame, str(box['player_id']), (x1, y2 +45), cv2.FONT_HERSHEY_PLAIN, 4, (255, 255, 255), 3)
+
+            # for each frame the event data is checked
+            valid_event_rows = event.dropna(subset=['x1', 'y1', 'x2', 'y2'])
+            for index, row in valid_event_rows.iterrows():
+             # if the data matches the current frame it is checked for if it is an error or not it is then given a green or red box based off this information
+              if row['frame_id'] == frame_index + 1:
+                if row['error'] == 'Yes':
+                  cv2.rectangle(frame, (int(row['x1']), int(row['y1'])), (int(row['x2']), int(row['y2'])), (0, 0, 255), 2)
+                  cv2.putText(frame, 'Error', (int(row['x1']), int(row['y1']) +45), cv2.FONT_HERSHEY_PLAIN, 4, (0, 0, 255), 3)
+                else:
+                  cv2.rectangle(frame, (int(row['x1']), int(row['y1'])), (int(row['x2']), int(row['y2'])), (0, 255, 0), 2)
+                  cv2.putText(frame, str(row['event_name']), (int(row['x1']), int(row['y1']) +45), cv2.FONT_HERSHEY_PLAIN, 4, (0, 255, 0), 3)
+                  cv2.putText(frame, 'player ID: '+ str(row['player_id']) + ' Event: ' + str(row['event_name']), (50, 150), cv2.FONT_HERSHEY_PLAIN, 4, (0, 255, 0), 3)
+
+
+    # writing to the new output video
+    out.write(frame)
+
+    if cv2.waitKey(1) & 0xFF == ord('q'):
+        break
+
+    frame_index += 1
+
+cap.release()
+out.release()
+cv2.destroyAllWindows()

26_T1/afl_player_tracking_and_crowd_monitoring/Crowd_Monitoring/Crowd_Alert_and_Forecasting/Forecasting_Model/README.md

@@ -0,0 +1,116 @@
+# Forecasting Model Documentation
+
+## Overview
+This document provides a detailed overview of the forecasting model developed for predicting the number of visitors to a place. The model uses the **Prophet algorithm** and incorporates features such as the **day of the week** and **user-specified forecasting periods**. The following sections cover installation, usage, and output details, including placeholders for screenshots.
+---
+## 1. Installation
+
+### 1.1 Requirements
+To run the forecasting model, ensure the following packages are installed:
+- pandas
+- prophet
+- matplotlib
+- openpyxl (for saving output to Excel)
+
+### 1.2 Install the Required Packages
+Before running the visitor forecasting model, ensure you have the following Python packages installed in your environment. These libraries are essential for data handling, forecasting, plotting graphs, and saving results.
+
+#### Step-by-Step Installation:
+You can install these packages using `pip` by running the following commands in your terminal or command prompt:
+
+1. Install pandas:
+   ```bash
+   pip install pandas
+   ``` 
+
+2. Install Prophet (this will also install cmdstanpy, which Prophet depends on):
+   ```bash
+   pip install prophet
+   ```
+
+3. Install matplotlib:
+   ```bash
+   pip install matplotlib
+   ```
+
+4. Install openpyxl (for exporting results to Excel):
+   ```bash
+   pip install openpyxl
+   ```
+
+Alternatively, you can install all the required packages at once by running:
+   ```bash
+    pip install pandas prophet matplotlib openpyxl
+   ```
+
+#### Verifying the Installation: 
+Once the packages are installed, you can verify them by importing them in Python. Open a Python environment (either a Jupyter Notebook or Python shell) and run:
+
+```
+import pandas as pd
+from prophet import Prophet
+import matplotlib.pyplot as plt
+import openpyxl
+```
+If there are no errors, the packages are installed correctly, and you are ready to run the forecasting model.
+
+---
+## 2. How to Use the Forecasting Model
+
+### 2.1 Load the Data
+- Prepare a CSV file containing historical visitor data, with columns for date and visitor count.
+- For demonstration, website visitor data was used to build the model.
+
+### 2.2 Execute the Script
+- Run the provided script to start the forecasting process.
+- You will be prompted to input the number of days to forecast.
+
+---
+
+## 3. User Interaction
+
+### 3.1 Input Prompt for Number of Days
+- Once the script is executed, it will prompt you to input a number representing how many days into the future you want to predict.
+- Enter a valid numerical value.
+```
+How many days do you want to forecast? [Input Box]
+```
+![Input Prompt Example](./images/input.png)
+
+---
+
+## 4. Output Details
+
+### 4.1 Forecast Graph
+- After execution, the model will display a forecast graph:
+  - **X-axis**: Date
+  - **Y-axis**: Predicted number of visitors
+- The graph will include both historical data and the forecast for easy comparison.
+
+![Forecast Graph Example](./images/forecast_graph.png)
+
+---
+
+## 5. Saving the Results
+
+### 5.1 Exporting to Excel
+- The forecasted visitor data (with lower and upper confidence intervals) is automatically saved in an Excel file named `visitor_forecast_output.xlsx`.
+- This file can be used for further analysis or sharing.
+
+![User Input Example](./images/user_input_example.png)
+
+---
+
+## 6. Technical Details
+
+### 6.1 Data Preprocessing
+- The historical visitor data is processed by:
+  - Converting the **Date** column into a `datetime` object.
+  - Cleaning the **Visitor.Count** column by removing any non-numerical characters to ensure accurate numerical operations.
+
+### 6.2 Model Features
+- **Days of the Week**: The model accounts for variations in visitor patterns on weekdays vs. weekends.
+- **Holidays (Optional)**: For greater accuracy, the model can optionally incorporate holiday data, which is particularly useful during special events or holiday seasons.
+
+---
+