import matplotlib.pyplot as plt import numpy as np import tensorflow as tf from sklearn import datasets from tensorflow.python.framework import opsnode
ops.reset_default_graph()python
sess = tf.Session()app
irid = datasets.load_iris() x_vals = np.array([x[0:3] for x in irid.data]) y_vals = np.array([x[3] for x in irid.data])dom
seed = 2 tf.set_random_seed(seed) np.random.seed(seed)code
train_indices = np.random.choice(len(x_vals), round(len(x_vals) * 0.8), replace=False) test_indices = np.array(list(set(range(len(x_vals))) - set(train_indices))) x_vals_train = x_vals[train_indices] x_vals_test = x_vals[test_indices]orm
y_vals_train = y_vals[train_indices] y_vals_test = y_vals[test_indices]get
def normalize_cols(m): col_max = m.max(axis = 0) col_min = m.min(axis = 0) return (m-col_min)/(col_max-col_min)it
x_vals_train = np.nan_to_num(normalize_cols(x_vals_train)) x_vals_test = np.nan_to_num(normalize_cols(x_vals_test))io
batch_size = 50 x_data = tf.placeholder(shape=[None,3],dtype=tf.float32) y_target = tf.placeholder(shape=[None,1],dtype=tf.float32)test
hidden_layer_nodes = 10 A1 = tf.Variable(tf.random_normal(shape=[3,hidden_layer_nodes])) b1 = tf.Variable(tf.random_normal(shape=[hidden_layer_nodes]))
B1 = tf.Variable(tf.random_normal(shape=[hidden_layer_nodes,1])) b2 = tf.Variable(tf.random_normal(shape=[1]))
hidden_output = tf.nn.relu(tf.add(tf.matmul(x_data,A1),b1)) final_output = tf.nn.relu(tf.add(tf.matmul(hidden_output,B1),b2))
loss = tf.reduce_mean(tf.square(y_target-final_output))
my_opt = tf.train.GradientDescentOptimizer(0.005) train_step =my_opt.minimize(loss)
init = tf.global_variables_initializer() sess.run(init)
loss_vec = [] test_loss = []
for i in range(500): rand_index = np.random.choice(len(x_vals_train), size=batch_size) rand_x = x_vals_train[rand_index] rand_y = np.transpose([y_vals_train[rand_index]]) sess.run(train_step, feed_dict={x_data: rand_x, y_target: rand_y})
temp_loss = sess.run(loss, feed_dict={x_data: rand_x, y_target: rand_y}) loss_vec.append(np.sqrt(temp_loss)) test_temp_loss = sess.run(loss, feed_dict={x_data: x_vals_test, y_target: np.transpose([y_vals_test])}) test_loss.append(np.sqrt(test_temp_loss))
plt.plot(loss_vec, 'k-', label='Train Loss') plt.plot(test_loss, 'r--', label='Test Loss') plt.title('Loss (MSE) per Generation') plt.legend(loc='upper right') plt.xlabel('Generation') plt.ylabel('Loss') plt.show()