PiTrac Range Session – Shot & Image Viewer

DISCLAIMER: This project relies heavily on LLM generated code and lacks so far sufficient testing and verification. Please verifiy data quality by randomly comparing the raw log and image files to the dasboard visualisation.

This project processes PiTrac shot data by combining the log file and images trough a timestamp matching logic to create a interactive analysis tool.
It loads shot results from the log file, matches it with captured images, exports the combined dataset as a .pkl database, and provides an interactive dashboard for exploring each shot in detail.

---

Features

Logfile Parsing

Reads raw .log files containing shot timestamps, ball data, and CSV‑based metrics. Can be drawn from:

/home/user/.pitrac/logs

Image Loading

Loads all images from directory and extracts timestamps and metadata. Can be drawn from:

/home/user/LM_Shares/Images

Shot–Image Matching

Automatically matches each shot with the closest images (within a 1‑minute window).

PKL + Excel Export

Saves the merged dataset as:

• a .pkl file for fast re‑loading
• an optional .xlsx export for external analysis and troubleshooting

Interactive Dash Plot

A browser‑based dashboard that includes:

• scatter plot (configurable y-content vs. shot time-stamp)
• hover‑based shot selection
• keyboard navigation (← / →) -> not working right now
• automatic image display for each shot
• grouped spin images and other images
• special ordering (e.g., final_found_ball always last)

---

Project Structure

pitrac_rangesession/
│
├── src/
│   ├── log_parser.py
│   ├── image_loader.py
│   ├── matcher.py
│   └── plotting.py
├── data/
│   ├── example/ #minimal example if you have no data at hand
│   │   ├── images/ #example images from three shots
│   │   ├── sample.log # three shots from wich one was not detected
│   ├── test_*.log #drop your logfile here
│   └── images/ #drop your images here
├── output/ 
│   └── data.pkl  #will be created after you run main.py
├── exports/ 
│   └── data.xlsx  #will be created after you run main.py
├── .venv/ 
├── excel_export.py 
├── .gitignore
├── main.py 
├── pyproject.toml
└── README.md

---

How It Works (main.py)

1. Parse Logfile

Extracts shot timestamps, ball speed, club speed, spin, and other metrics.

2. Load Images

Reads all images from the session folder and extracts timestamps from filenames.

3. Match Shots ↔ Images

Each shot is matched with the closest images based on timestamp proximity.

4. Save as PKL

The merged dataset is stored as output/data.pkl.

5. Optional Excel Export

The PKL is exported to exports/data.xlsxy for easy verification

6. Interactive Plot

A Dash application opens in your browser and allows:

• exploring shots visually by hovering over datapoints in the diagram
• comparing metrics
• viewing matched images

---

Running the Project

Activate your virtual environment and run:

python main.py