Skip to content

Latest commit

 

History

History
87 lines (70 loc) · 1.19 KB

README.md

File metadata and controls

87 lines (70 loc) · 1.19 KB
Raw

Multilayer Perceptron

Multilayer Perceptron based on NumPy

Using Backpropagation with Stochastic gradient descent to optimize network.

Layers

  • InputLayer
  • TanhLayer
  • SigmoidLayer
  • ReLULayer
  • LeakyReLULayer
  • DropoutLayer

See: nn.py

Examples

See:

  1. Basic example
  2. Auto encoder example
  3. Auto encoder & dropout example

Auto encoder performance of this example

input:
    ■
  ■ ■ ■ ■
■   ■   ■
    ■
    ■

prediction:
    ■
  ■ ■ ■
■   ■   ■
    ■
    ■

---------
input:
    ■
      ■
■ ■ ■ ■ ■
      ■ ■
    ■

prediction:
    ■
      ■
■ ■ ■ ■ ■
      ■
    ■

---------
input:
    ■
    ■
■   ■   ■
■ ■ ■ ■
    ■

prediction:
    ■
    ■
■   ■   ■
  ■ ■ ■
    ■

---------
input:
    ■
  ■
■ ■ ■ ■ ■
  ■
■   ■

prediction:
    ■
  ■
■ ■ ■ ■ ■
  ■
    ■