参考官方文档感觉这两个函数作用差不多但是我习惯用tf.argmax
format:argmax(input, axis=None, name=None, dimension=None)
Args:input: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`.
axis: A `Tensor`. Must be one of the following types: `int32`, `int64`.(注意这里是整形就行了)
int32, 0 <= axis < rank(input). Describes which axis
of the input Tensor to reduce across. For vectors, use axis = 0.
name: A name for the operation (optional).
栗子
import tensorflow as tf
x=tf.constant([[1.,2.,6],[6.,2.,6]])
xShape=tf.shape(x)
z1=tf.arg_max(x,1)#沿axis=0操作
with tf.Session() as sess:
xShapeValue,d1=sess.run([xShape,z1])
print('shape= %s'%(xShapeValue))
print(d1)
结果
一般用来求一个矩阵中,每行最大值的index,一般用在测试的时候。