代码参考自: Tensorflow官方微信公众号
1.Tensorflow下用都是tf 写的格式,数据也是,所以如果来自pytorch的话需要转化一下,比如
:
可以运行
cons_tensor = tf.constant([[1,2],[3,4]],dtype=tf.int32)
不可以运行
cons_tensor = tf.constant([[1,2],[3,4]],dtype=int32)
2.简单的梯度下降加上线性回归:
X_raw = np.array([2013,2014,2015,2016,2017],dtype=np.float32)
Y_raw = np.array([12000,14000,15000,16000,175000],dtype=np.float32)
X = (X_raw - X_raw.min()) / (X_raw.max()- X_raw.min())
Y = (Y_raw - Y_raw.min()) / (Y_raw.max()- Y_raw.min())
X = tf.constant(X)
Y = tf.constant(Y)
a = tf.Variable(initial_value=0.)
b = tf.Variable(initial_value=0.)
variables = [a,b]
num_epoch =10000
optimizer = tf.keras.optimizers.SGD(learning_rate=1e-3)
for e in range(num_epoch):
with tf.GradientTape() as tape:
y_pred = a*