Hog Data Tool

A Python project for loading and visualizing HOG grip training session data.

current capabilities

• load hog data into a structured format for easy access and analysis StructuredHogData.
• Easy access grouped Session data (i.e micro right hand) with some additional computed features, FullSessionData
• Filtering of sessions (i.e include sessions 10 to current) allowing for some simple visualisations pre hatching.
• bulk plotting methods across all gripper types and hands.
• basic curve fitting (hyperbolic)
• simple set of visualisations
  • power and inverted power curves
  • next weight suggestions based on least dense regions of power curve
  • session frequency and session gaps
  • rolling average weight of each regieme (power, strength, endurance etc)

---

Setup

1. Environment

   Copy env.shared to .env.local and update paths and settings:

   INPUT_DATA_PATH=./data/hog_data.csv # path to hog data
   OUTPUT_DATA_PATH=./data/outputs/ # path output plotd saved to
   WEIGHT_UNIT=lbs #lbs or kgs
   
   

2. Install dependencies

   simply call poetry install and poetry will install the dependencies to your venv.

---

Running the project

Run the main data pipeline make run

This will

1. Load the CSV from INPUT_DATA_PATH.

2. Create plots for all grippers:

   • Power curves (plot_power_curve)
   • Inverted power curves (plot_inverted_power_curve)
   • Session gaps (plot_session_gap)
   • Session frequency (plot_session_frequency)
   • Rolling average weight in regimes (plot_rolling_average_weight_in_regimes)

3. Save outputs to OUTPUT_DATA_PATH.

---

Loading data

Data can be loaded directly in Python:

from hog_data_tool import StructuredHogData, EnvConfig, get_env_config

config = get_env_config()
data = StructuredHogData.from_csv(config.input_data_path)

---

Plotting

You can generate plots programmatically:

# Plot per gripper
data.create_plot_for_all_grippers(plot_method=plot_power_curve, output_path=config.output_data_path / "power_curve")

# Plot shared data (all right hands by default)
data.create_shared_gripper_plot(plot_method=plot_session_gap, output_path=config.output_data_path / "session_gap")

You can create your own plotting functions, note all plotting functions must follow the SessionPlotMethod or SharedSessionPlotMethod signature. """

---

Curve fitting

• Basic hyperbolic curve fitting is included (fit_power_curve_with_hyperbolic_decay).
• Piecewise curve fitting will be added in future updates.

---

Example plots