tensorflow - 如何使用反向传播/梯度节点生成和保存模型文件？

Question

关于生成和保存具有如下梯度节点的 tensorflow 模型以及 tensorflow repo 上的示例文件的任何想法： https://github.com/tensorflow/tensorflow/blob/master/tensorflow/compiler/jit/tests/keras_imagenet_main。 pbtxt

node {
name: “training/LossScaleOptimizer/gradients/res5c_branch2b_1/Conv2D_grad/Conv2DBackpropInput”
op: “Conv2DBackpropInput”
input: “ConstantFolding/training/LossScaleOptimizer/gradients/res5c_branch2b_1/Conv2D_grad/ShapeN-matshapes-0”
input: “res5c_branch2b_1/Conv2D/Cast”
input: “training/LossScaleOptimizer/gradients/bn5c_branch2b_1/FusedBatchNormV2_grad/FusedBatchNormGradV2”
device: “/job:localhost/replica:0/task:0/device:GPU:0”
attr {
key: “T”
value {
type: DT_HALF
}
}
attr {
key: “data_format”
value {
s: “NCHW”
}
}
attr {
key: “dilations”
value {
list {
i: 1
i: 1
i: 1
i: 1
}
}
}

标签： tensorflowbackpropagation