MoonBoard

This repo is an experimental for utilizing Machine Learning technology for Rock Climbing training board named MoonBoard.

Directory organization

datasets/                # Dataset from MoonBoard
  +- create_moonboard.py # Create moonboard.npz from climbs.txt/grades.txt
  +- climbs.txt          # Dataset from michaelplesser/moonboard-NN
  +- grades.txt          # Dataset from michaelplesser/moonboard-NN
  +- load_moonboard.py   # Data loader for MoonBoard
  +- moonboard.npz       # Training and Test datasets for MoonBoard

Usage

Loading dataset from Python

• Its usage is quite similar to keras.datasets.mnist.load_data() function.
• The argument represents the path to moonboard.npz to be loaded.
• The return values are same as keras.datasets.mnist.load_data().

>>> from datasets.load_moonboard import load_moonboard
>>> (x_train, y_train), (x_test, y_test) = load_moonboard('./datasets/moonboard.npz')
>>> print(x_train[0])
  [[0 0 0 0 0 0 0 1 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 1 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 1 0 0 0 0 0 0 0 1]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 1 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 1 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]
  [0 0 0 0 0 0 0 0 0 0 0]]
>>> print(y_train[0])
  10

Using with TensorFlow

• TBA

Contribution

References