Skip to content

Latest commit

 

History

History
12 lines (10 loc) · 527 Bytes

README.md

File metadata and controls

12 lines (10 loc) · 527 Bytes

ML-JS

Machine learning in javascript without any external machine learning library just maths.

Examples

  • XOR (Trained using backpropogration)
  • Flappy bird (Trained using neuro evolution)

NOTE

  • The prediction might be wrong because of incorrect weights and biases.
  • Weights and biases can be fixed by trainning more.
  • Accuracy of the model can be increased be increasing number of hidden layers and the neurons in them.
  • Higher number of layers and neuron need more training to reach correct weights and biases.