######### Tensorflow常用函数 ##########
目录
算数运算
基本运算
tf.add(x, y, name=None) # 求和
tf.sub(x, y, name=None) # 减法
tf.mul(x, y, name=None) # 乘法
tf.div(x, y, name=None) # 除法
tf.mod(x, y, name=None) # 取模
tf.abs(x, name=None) # 求绝对值
tf.neg(x, name=None) # 取负
函数
tf.square(x, name=None) # 计算平方
tf.round(x, name=None) # 舍入最接近的整数 [0.9, 2.5, 2.3, -4.4] > [ 1.0, 3.0, 2.0, -4.0 ]
tf.sqrt(x, name=None) # 开根号
tf.pow(x, y, name=None) # 幂次方
tf.exp(x, name=None) # 计算e的次方
tf.log(x, name=None) # 计算log，一个输入计算e的ln，两输入以第二输入为底
tf.cos(x, name=None) #三角函数cosine
tf.sin(x, name=None) #三角函数sine
tf.tan(x, name=None) #三角函数tan
tf.atan(x, name=None) #三角函数ctan
返回最值
tf.maximum(x, y, name=None) #返回最大值
tf.minimum(x, y, name=None) #返回最小值
矩阵数学函数
tf.diag(diagonal, name=None)
tf.transpose(a, perm=None, name='transpose')
tf.matmul(a, b, transpose_a=False, transpose_b=False, a_is_sparse=False, b_is_sparse=False, name=None)
tf.batch_matmul(x, y, adj_x=None, adj_y=None, name=None)
tf.matrix_determinant(input, name=None)
tf.batch_matrix_determinant(input, name=None)
tf.matrix_inverse(input, name=None)
tf.batch_matrix_inverse(input, name=None)
tf.cholesky(input, name=None)
tf.batch_cholesky(input, name=None)
张量操作
创建张量
a= tf.constant([[1,5]],dtype=tf.int32) # 维度：一维为个数，二维用[行,列]，多维用[n,m,j...]
恒值张量
tf.zeros([2,3,2])
tf.ones(2)
tf.fill(2,3) # tf.fill(维度,指定值)
张量降维
tf.reduce_sum(input_tensor, reduction_indices=None, keep_dims=False, name=None)
tf.reduce_prod(input_tensor, reduction_indices=None, keep_dims=False, name=None)
tf.reduce_min(input_tensor, reduction_indices=None, keep_dims=False, name=None)
tf.reduce_max(input_tensor, reduction_indices=None, keep_dims=False, name=None)
tf.reduce_mean(input_tensor, reduction_indices=None, keep_dims=False, name=None)
tf.reduce_all(input_tensor, reduction_indices=None, keep_dims=False, name=None)
tf.reduce_any(input_tensor, reduction_indices=None, keep_dims=False, name=None)
tf.accumulate_n(inputs, shape=None, tensor_dtype=None, name=None)
序列
tf.linspace(start, stop, num, name=None) # 等差数列
tf.range(start, limit, delta=1) # delta为扩增量
x,y= np.mgrid[起始值:结束值步长, 起始值:结束值步长....] # 生成等差数组
x.ravel() # 将数组变为一维数组
np.c_[数组1, 数组2....] # 数组数值点配对
随机张量
tf.set_random_seed(seed)
tf.random_normal(shape, mean=0.0, stddev=1.0, dtype=tf.float32, seed=None, name=None)
tf.truncated_normal(shape, mean=0.0, stddev=1.0, dtype=tf.float32, seed=None, name=None)
tf.random_uniform(shape, minval=0.0, maxval=1.0, dtype=tf.float32, seed=None, name=None)
tf.random_shuffle(value, seed=None, name=None)
np.random.RandomState.rand(维度) # 默认返回标量 # 返回[0,1)之间的随机数
np.random.shuffle(x) # 将数据打乱, 可设置seed来确保数据和标签打乱后相对应
数据格式转化
tf.cast(x, dtype, name=None) # tf.cast(a, tf.int32)
模型搭建、训练、测试、预测常用函数
输入特征和标签值一一对应。将数据集分为batch组数据，分批次喂入神经网络
tf.data.Dataset.from_tensor_slices((x_train, y_train)).batch(32)
tf.Variable(tf.random.truncated_normal([3], stddev=0.1, seed=1)) # 标记参数可训练，在神经网络八股中不常用
tf.GradientTape() # 进行先计算模型的梯度(也就是偏导)
神经网络八股六步法
1 import
2 train, test数据
3 描述网络的各方面性质(搭建神经元结构)
model = tf.keras.models.Sequential([网络结构])
①拉直层
tf.keras.layers.Flatten()
②全连接层
tf.keras.layers.Dense(神经元个数, activation = "激活函数”, kernel_regularizer = 哪种正则化)
#其中的activation可选relu、softmax、sigmoid、tanh
#其中的kernel_regularizer可选: tf.keras.regularizers.l1()、tf.keras,regularizers.l2()
③卷积层
tf.keras.layer.Conv2D(filters = 卷积核个数, kernel_size = 卷积核尺寸, strides = 卷积步长， padding = "valid" or“same")
④LSTM层
tf.keras.layers.LSTM()
4 配置训练方法
model.compile(optimizer = 优化器,
loss = 损失函数,
metrics = ['准确率']) # 配置训练方法
其中metrics可选
'accuracy’: y和y_都是数值
'categorical_accuracy': y和y_都是独热码（概率分布）
'sparse_categorical_accuracy': 其中一个是数值，另一个是独热码
5 执行训练过程
model.fit(训练集的输入特征,训练集的标签,
batch_size= , epochs = ,
validation_data = =(测试集的输入特征，测试集的标签)
validation_split = ,
validation_freq = 多少次epoch测试一次)
6 打印
model.summary
八股class，用class搭建神经网络
class MyModel(Model):
def __init__(self):
super(MyModel,self).__init__()
self.name = 网络结构
def call(self,x):
y = self.name(x)
return y
model = MyMdoel()
保存模型或加载本地模型
checkpoint_save_path = "./checkpoint/mnist.ckpt"
if os.path.exists(checkpoint_save_path + '.index'):
print('-------------load the model-----------------')
model.load_weights(checkpoint_save_path)
# 对已存在的模型只读取最优参数
cp_callback = tf.keras.callbacks.ModelCheckpoint(filepath=checkpoint_save_path,
save_weights_only=True,
save_best_only=True)
# 模型执行训练过程
model.fit(x_train, y_train, batch_size=32, epochs=5, validation_data=(x_test, y_test), validation_freq=1,
callbacks=[cp_callback])
用已训练的模型进行预测
# 先搭建好神经元网络，再加载本地模型参数
model.load_weights(model_save_path)
# 先对数据进行预处理，再进行模型预测
result = model.predict(x_predict)
sequential和class源代码
sequential
import tensorflow as tf
from sklearn import datasets
import numpy as np
x_train = datasets.load_iris().data
y_train = datasets.load_iris().target
np.random.seed(116)
np.random.shuffle(x_train)
np.random.seed(116)
np.random.shuffle(y_train)
tf.random.set_seed(116)
model = tf.keras.models.Sequential([
tf.keras.layers.Dense(3, activation='softmax', kernel_regularizer=tf.keras.regularizers.l2())
])
model.compile(optimizer=tf.keras.optimizers.SGD(lr=0.1),
loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),
#这里注意，如果神经网络预测结果输出前没有经过概率分布，则是True，要是经过了概率分布，则是False，最典型的概率分布就是softmax
metrics=['sparse_categorical_accuracy'])
model.fit(x_train, y_train, batch_size=32, epochs=500, validation_split=0.2, validation_freq=20)
model.summary()
class
import tensorflow as tf
from tensorflow.keras.layers import Dense
from tensorflow.keras import Model
from sklearn import datasets
import numpy as np
x_train = datasets.load_iris().data
y_train = datasets.load_iris().target
np.random.seed(116)
np.random.shuffle(x_train)
np.random.seed(116)
np.random.shuffle(y_train)
tf.random.set_seed(116)
class IrisModel(Model):
def __init__(self):
super(IrisModel, self).__init__()
self.d1 = Dense(3, activation='softmax', kernel_regularizer=tf.keras.regularizers.l2())
def call(self, x):
y = self.d1(x)
return y
model = IrisModel()
model.compile(optimizer=tf.keras.optimizers.SGD(lr=0.1),
loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),
metrics=['sparse_categorical_accuracy'])
model.fit(x_train, y_train, batch_size=32, epochs=500, validation_split=0.2, validation_freq=20)
model.summary()
模型续训、保存、加载、预测源代码
模型训练
import tensorflow as tf
import os
import numpy as np
from matplotlib import pyplot as plt
# 设置打印的条数，无限条
np.set_printoptions(threshold=np.inf)
# 加载数据
mnist = tf.keras.datasets.mnist
(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0
# 搭建神经元网络结构
model = tf.keras.models.Sequential([
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(128, activation='relu'),
tf.keras.layers.Dense(10, activation='softmax')
])
# 配置训练方法
model.compile(optimizer='adam',
loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),
metrics=['sparse_categorical_accuracy'])
# 保存模型或加载本地模型
checkpoint_save_path = "./checkpoint/mnist.ckpt"
if os.path.exists(checkpoint_save_path + '.index'):
print('-------------load the model-----------------')
model.load_weights(checkpoint_save_path)
# 对已存在的模型只读取最优参数
cp_callback = tf.keras.callbacks.ModelCheckpoint(filepath=checkpoint_save_path,
save_weights_only=True,
save_best_only=True)
# 模型执行训练过程
history = model.fit(x_train, y_train, batch_size=32, epochs=5, validation_data=(x_test, y_test), validation_freq=1,
callbacks=[cp_callback])
model.summary()
print(len(y_train),len(y_test))
######################### 打印和存储参数 ###############################################
# print(model.trainable_variables)
# file = open('./weights.txt', 'w')
# for v in model.trainable_variables:
# file.write(str(v.name) + '\n')
# file.write(str(v.shape) + '\n')
# file.write(str(v.numpy()) + '\n')
# file.close()
############################################### show ###############################################
# 显示训练集和验证集的acc和loss曲线
acc = history.history['sparse_categorical_accuracy']
val_acc = history.history['val_sparse_categorical_accuracy']
loss = history.history['loss']
val_loss = history.history['val_loss']
plt.subplot(1, 2, 1)
plt.plot(acc, label='Training Accuracy')
plt.plot(val_acc, label='Validation Accuracy')
plt.title('Training and Validation Accuracy')
plt.legend()
plt.subplot(1, 2, 2)
plt.plot(loss, label='Training Loss')
plt.plot(val_loss, label='Validation Loss')
plt.title('Training and Validation Loss')
plt.legend()
plt.show()
模型预测
from PIL import Image
import numpy as np
import tensorflow as tf
model_save_path = './checkpoint/mnist.ckpt'
model = tf.keras.models.Sequential([
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(128, activation='relu'),
tf.keras.layers.Dense(10, activation='softmax')])
model.load_weights(model_save_path)
preNum = int(input("input the number of test pictures:"))
for i in range(preNum):
image_path = input("the path of test picture:")
img = Image.open(image_path)
img = img.resize((28, 28), Image.ANTIALIAS)
img_arr = np.array(img.convert('L'))
img_arr = 255 - img_arr
img_arr = img_arr / 255.0
print("img_arr:",img_arr.shape)
x_predict = img_arr[tf.newaxis, ...]
print("x_predict:",x_predict.shape)
result = model.predict(x_predict)
pred = tf.argmax(result, axis=1)
print('\n')
tf.print(pred)
扫一扫
1015
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
