def train(request):
print("begin")
training_data, validation_data, test_data = mnl.load_data_wrapper()
print("loaded ... starting training")
ar,steps = net.SGD(training_data, 4, 10, 3.0, test_data=test_data)
data = {'error':ar,'batch':steps}
return JsonResponse(data)
import network.network as Network
import network.mnist_loader as mnist_loader
import pickle
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import axes3d, Axes3D
import numpy as np
from ripser import ripser
from persim import plot_diagrams
with open('network/trained_network.pkl', 'rb') as f:
u = pickle._Unpickler(f)
u.encoding = 'latin1'
net = u.load()
train_data, train_labels, test_data, test_labels = mnist_loader.load_data_wrapper(
)
def predict(n):
# Get the data from the test set
x = test_data[n]
# Print the prediction of the network
print('Network output: \n' + str(np.round(net.feedforward(x), 2)) + '\n')
print('Network prediction: ' + str(np.argmax(net.feedforward(x))) + '\n')
print('Actual image: ')
# Draw the image
plt.figure()
plt.imshow(x.reshape((28, 28)), cmap='Greys')
plt.show()
import matplotlib.pyplot as plt
import numpy as np
#with open('network/trained_network.pkl', 'rb') as f:
# net = cPickle.load(f, encoding = 'latin1')
# PYTHON 3 WORK AROUND (uncomment this
# and comment the above if using python 3)
with open('network/trained_network.pkl', 'rb') as f:
u = pickle._Unpickler(f)
u.encoding = 'latin1'
net = u.load()
#net = cPickle.load(f, encoding = 'latin1')
training_data, validation_data, test_data = mnist_loader.load_data_wrapper()
def predict(n):
# Get the data from the test set
x = test_data[n][0]
# Print the prediction of the network
print('Network output: \n' + str(np.round(net.feedforward(x), 2)) + '\n')
print('Network prediction: ' + str(np.argmax(net.feedforward(x))) + '\n')
print('Actual image: ')
# Draw the image
plt.imshow(x.reshape((28,28)), cmap='Greys')
# Replace the argument with any number between 0 and 9999
#predict(8384