def main():
X, C, X_tst, C_tst = load_isolet()
#print(X)
Y = np.array([[1, 2, 3, 4]])
print(X.shape)
#print(C)
#print(C.shape)
b1 = BatchNormalizer(X)
b1.getMean()
b1.getStd()
b1.getNormalized()
b1.getBatches()
def main():
X, C, X_tst, C_tst = load_isolet()
#print(X)
Y = np.array([[1, 2, 3, 4]])
#print(X.shape)
#print(C)
#print(C.shape)
b1 = BatchNormalizer(X, C, batch_size=40, shuffle=True)
train = b1.getBatches(X, C)
test = b1.getBatches(X_tst, C_tst, test=True)
#print(b1.cbatches.batch_num)
print(X.shape)
print(X_tst.shape)
from Trainer import Trainer
from Evaluator import Evaluator
from Model import Model
from nn18_ex2_load import load_isolet
import numpy as np
import logging
import os
if __name__ == '__main__':
# training and validation error collector
ec = ErrorCollector()
X, C, X_tst, C_tst = load_isolet()
#print("X shape: ", X.shape[0])
#print("X shape: ", X.shape[1])
# Parameters and model
epochs = 5
#learning_rate = 0.001
learning_rate = 0.01
model = Model(n_in=X.shape[1], n_hidden=300, n_out=26, n_layer=1)
batch_size = 40
# setup logging
log_file_name = 'logs' + os.sep + 'Log' + '_ep' + str(
epochs) + '_hidu' + str(model.n_hidden) + '_hidl' + str(
model.n_layer) + '_lr' + str(learning_rate) + '.log'
logging.basicConfig(filename=log_file_name, level=logging.INFO)
import tensorflow as tf
import numpy as np
import numpy.random as rd
from nn18_ex2_load import load_isolet
import matplotlib.pyplot as plt
Xa, C, X_test, C_test = load_isolet()
print(Xa.shape)
print(C.shape)
print(X_test.shape)
print(C_test.shape)
#We create the C arrays with this function
def createC(array):
res = np.zeros((len(array), 26))
for i in range(len(array)):
res[i][array[i] - 1] = 1
return res
Ca = createC(C)
Ca_Test = createC(C_test)
# Give the dimension of the data and chose the number of hidden layer
n_in = 300
n_out = 26
n_hidden = 20
# Set the variables
W_hid = tf.Variable(rd.randn(n_in, n_hidden) / np.sqrt(n_in), trainable=True)
