Skip to content

Chancing0/EZ-BP-MLP-Mnist

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 

Repository files navigation

BP_Mnist_ezCode_MLP

ezCode of Backpropagation with sigle hidden layer to train Mnist and some Gate.
Please pip install sklearn matplotlib pylab numpy

Run mlp_DigitsMnist.py

just run , but you could change the model hyperparameters by below code line
def train(self, data, iterations=1000, alpha=0.01)
alpha: learning rate
iterations:The training times of the model you want.

Run mlp_Gate.py

  1. just change the data inside below
    def demo():
    data = np.array([
    [[0,0,0], [0]],
    [[0,1,0], [1]],
    [[1,0,0], [1]],
    [[1,1,1], [0]]
    ])
    as you can see , here is four samples, each samlpe has 3 inputs feature , and one lable.
    it's XOR gate Truth table with extended feture(X1*X2)
  2. change the hyperparameters baseon your data form of below code to built your NN
    n = NN(3, 10, 1)
    it means input layer with 3 node of feature . single hidden layer with 10 nodes, output layer with 1 output.
  3. just run , but you could change the model hyperparameters by below code line
    def train(self, data, iterations=1000, alpha=0.01):
    alpha: learning rate
    iterations:The training times of the model you want.

About

ezCode of Backpropagation with sigle hidden layer to train Mnist or Gate.

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages