Python–张量流。索引策略图属性
原文:https://www . geesforgeks . org/python-tensorflow-indexed slices-graph-attribute/
TensorFlow 是谷歌设计的开源 Python 库，用于开发机器学习模型和深度学习神经网络。
图形用于查找包含值、索引和形状张量的图形。
语法:张量流。索引策略.图表
返回:它返回一个图形实例。
例 1:
Python 3
# Importing the library
import tensorflow as tf
# Initializing the input
data = tf.constant([[1, 2, 3], [4, 5, 6]], dtype = tf.float32)
# Printing the input
print('data: ', data)
# Calculating result
res = tf.IndexedSlices(data, [0])
# Finding Graph
@tf.function
def gfg():
tf.compat.v1.disable_eager_execution()
graph = res.graph
# Printing the result
print('graph: ', graph)
gfg()
输出:
data: Tensor("Const_1:0", shape=(2, 3), dtype=float32)
graph: <tensorflow.python.framework.ops.Graph object at 0x7f2eeda9e630>
<tf.Operation 'PartitionedCall_1' type=PartitionedCall>
例 2:
Python 3
# Importing the library
import tensorflow as tf
# Initializing the input
data = tf.constant([1, 2, 3])
# Printing the input
print('data: ', data)
# Calculating result
res = tf.IndexedSlices(data, [0])
# Finding Graph
graph = res.graph
# Printing the result
print('graph: ', graph)
输出:
data: Tensor("Const_6:0", shape=(3, ), dtype=int32)
graph: <tensorflow.python.framework.ops.Graph object at 0x7f2eeda9e630>
版权属于：月萌API www.moonapi.com，转载请注明出处
