def main():
if len(sys.argv) < 2:
print('must specify 2 files')
print(_f('usage: {sys.argv[0]} <file1> <file2>'))
return
f1, f2 = open(sys.argv[1]), open(sys.argv[2])
compareFiles(f1, f2, [lambda a,b: b-a, lambda a,b: (a-b)/b])
def get_filename(self):
fname = filedialog.askopenfilename(title='quantify - Open')
print(_f("Opening {fname}"))
for idx in range(int(np.floor(train_size / batch_size))):
X = X_train[idx*batch_size:(idx+1)*batch_size]
Y = Y_train[idx*batch_size:(idx+1)*batch_size]
# Compute the gradient
w_grad, b_grad = u._gradient(X, Y, w, b)
# gradient descent update
# learning rate decay with time
w = w - learning_rate/np.sqrt(step) * w_grad
b = b - learning_rate/np.sqrt(step) * b_grad
step = step + 1
# Compute loss and accuracy of training set and development set
y_train_pred = u._f(X_train, w, b)
Y_train_pred = np.round(y_train_pred)
train_acc.append(u._accuracy(Y_train_pred, Y_train))
train_loss.append(u._cross_entropy_loss(y_train_pred, Y_train) / train_size)
y_dev_pred = u._f(X_dev, w, b)
Y_dev_pred = np.round(y_dev_pred)
dev_acc.append(u._accuracy(Y_dev_pred, Y_dev))
dev_loss.append(u._cross_entropy_loss(y_dev_pred, Y_dev) / dev_size)
print('Training loss: {}'.format(train_loss[-1]))
print('Development loss: {}'.format(dev_loss[-1]))
print('Training accuracy: {}'.format(train_acc[-1]))
print('Development accuracy: {}'.format(dev_acc[-1]))
# Plot
def window_callback(self, event):
print(_f('{event.widget} resized ({event.width}, {event.height})'))
return 'break'
