# KNN
def knn(train_x, train_y):
print('************* KNN ************')
# k的个数 uniform是均等的权重 distance是不均等的权重
model = KNeighborsClassifier(n_neighbors=4,weights='distance',n_jobs=1)
model.fit(train_x, train_y)
return model
if __name__ == '__main__':
# 加载数据集
print('reading training and testing data...')
train_x, train_y, test_x, test_y = mnist_load.get_data()
# 将标签二值化
# train_y = label_binarize(train_y,classes=[0,1,2,3,4,5,6,7,8,9])
# test_y = label_binarize(test_y,classes=[0,1,2,3,4,5,6,7,8,9])
# # 数据类型改为float32，单精度浮点数
# train_x = train_x.astype('float32')
# test_x = test_x.astype('float32')
# # 数据归一化
# train_x /= 255
# test_x /= 255
start_time = time.time()
# psum1 = 0
plt.figure(figsize=(n_cols * 1.4, n_rows * 1.6))
for row in range(n_rows):
for col in range(n_cols):
index = n_cols * row + col
plt.subplot(n_rows, n_cols, index + 1)
plt.imshow(x_data[index].reshape(28, 28),
cmap="binary",
interpolation='nearest')
plt.axis('off')
plt.title(class_name[y_data[index]])
plt.show()
return None
if __name__ == "__main__":
x_train, y_train, x_test, y_test = load.get_data()
show_imgs(7, 7, x_train, y_train, class_name) #展示功能
#不一样的是，神经网络进行浮点数计算，这里进行归一化
x_train = x_train.reshape((-1, 28, 28, 1)).astype("float32") / 255
x_test = x_test.reshape((-1, 28, 28, 1)).astype("float32") / 255
#建立模型
model = keras.Sequential([
#(-1,28,28,1)->(-1,28,28,32)
keras.layers.Conv2D(input_shape=(28, 28, 1),
filters=32,
kernel_size=5,
strides=1,
padding='same'), # Padding method),
#(-1,28,28,32)->(-1,14,14,32)