#!/usr/bin/env python3
# -*- coding: utf-8 -*-
#use evolutionFit to solve math problem
import numpy as np
from PSNN import model, layers
netmodel = model([layers.dense(1)])
inaccuracy = 1e-14
def mycubic(model):
# 2*x*x*x + 5*x*x = 10
x = model.predict([])[0]
loss = abs(10 - (2 * x * x * x + 5 * x * x))
return loss
netmodel.create(inputs=(0, ))
drate = 1
lastlost = 0
while True:
loss = netmodel.evolutionFit(rate=drate,
replication=200,
lossfunc=mycubic,
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
import numpy as np
from PSNN import model, layers, activation
netmodel = model([
layers.dense(2, activation.sigmoid()),
layers.dense(1, activation.sigmoid())
])
netmodel.debug = True
netmodel.create(inputs=(3, ))
inputs = np.array([
[0, 0, 1], #0
[0, 0, 0], #0
[1, 0, 0], #1
[1, 1, 1], #1
[0, 1, 0], #1
[0, 1, 1] #1
])
targets = np.array([
[0],
[0],
[1],
[1],
[1],
[1],
def getITArray(directory, target):
imgs = glob.glob(directory + "/*.png")
out = ([], [])
for loc in imgs:
img = Image.open(loc).convert('L') # convert image to 8-bit grayscale
WIDTH, HEIGHT = img.size
data = list(img.getdata()) # convert image data to a list of integers
out[0].append([data[offset:offset+WIDTH] for offset in range(0, WIDTH*HEIGHT, WIDTH)])
out[1].append(target)
return out
netmodel = model([
layers.flatten(),
layers.activation(activation.tanh()),
layers.dense(30, activation.tanh()),
layers.dense(20, activation.tanh()),
layers.dense(10, activation.sigmoid())
])
netmodel.debug = True
netmodel.create(
inputs=(28,28)
)
# get learn data from images
learn = {
"inputs": [],
"targets": []
}