import sys
sys.path.append("../..")
import lab0.src.dataset8 as dataset8
from sklearn import metrics
from neupy import algorithms
import numpy as np
from lib.image import noise, get_pxs
print(
"\n\nИсследуем качество классификации на тестовой выборке содержащей зашумленные примеры"
)
(x_train, y_train), (x_test, y_test) = dataset8.load_data(mode=0)
pnn = algorithms.PNN(std=1, batch_size=128, verbose=False)
pnn.train(x_train[0:10000], y_train[0:10000])
y_predicted = pnn.predict(x_test)
local_path = 'my_images/'
for nTest in np.arange(0, 10, 1):
# convert to numpy array
x = get_pxs(local_path + str(nTest) + '.png')
# Inverting and normalizing image
x = 255 - x
x /= 255
# 1,2 initializing
batch_size = 32
num_classes = 10
epochs = 10
lr = 0.01
verbose = 1
neurons_number = [256, num_classes]
opt_name = "Adam"
optimizer = Adam()
goal_loss = 0.013
(x_train, y_train), (x_test,
y_test) = dataset8.load_data(mode=2,
show=True,
show_indexes=[0, 1])
model = Sequential()
model.add(Dense(neurons_number[0], input_dim=28**2, activation='relu'))
model.add(Dense(neurons_number[1], activation='softmax'))
# 3 setting stopper
callbacks = [
EarlyStoppingByLossVal(monitor='val_loss', value=goal_loss, verbose=1)
]
# 4 model fitting
model.compile(optimizer=optimizer,