先把checkpoint模型转为saved model（这个过程中没报错），再把saved model转onnx报错了

DetectionTeamUCAS / FPN_Tensorflow

This is a tensorflow re-implementation of Feature Pyramid Networks for Object Detection.

https://github.com/DetectionTeamUCAS/FPN_Tensorflow

MIT License

347 stars 132 forks source link

先把checkpoint模型转为saved model（这个过程中没报错），再把saved model转onnx报错了 #151

Open HUI11126 opened 3 years ago

HUI11126 commented 3 years ago

saved model转onnx时，报错： ValueError: Node 'gradients/resnet_v1_101_4/block4/unit_3/bottleneck_v1/conv3/BatchNorm/FusedBatchNorm_grad/FusedBatchNormGrad' has an _output_shapes attribute inconsistent with the GraphDef for output #3: Dimension 0 in both shapes must be equal, but are 0 and 2048. Shapes are [0] and [2048].

请问，怎么解决？

HUI11126 commented 3 years ago

checkpoint模型转为saved model是在tensorflow1.14版本转的，saved model转onnx是在tensorflow 2.2版本转的

HUI11126 commented 3 years ago

若是用tensorflow 1.14转saved model为onnx，则内存快速被占满，之后被杀死

exmee commented 2 years ago

您好，请问怎么把checkpoint模型转为saved model呢，网络的输出节点output_node_names是什么呢，期待您的回复！