Froze graph to pb, Exception: 1 Unprocessed nodes

[Maple]

When i try to use the freezeTrainedBert.py to freeze the graph , i choose the node name : loss/SparseSoftmaxCrossEntropyWithLogits/SparseSoftmaxCrossEntropyWithLogits, but i got follow exception:

Blockquote
Exception in thread “main” java.lang.IllegalStateException: 1 Unprocessed nodes: [loss/SparseSoftmaxCrossEntropyWithLogits/SparseSoftmaxCrossEntropyWithLogits]
at org.nd4j.common.base.Preconditions.throwStateEx(Preconditions.java:641)
at org.nd4j.common.base.Preconditions.checkState(Preconditions.java:340)
at org.nd4j.imports.graphmapper.tf.TFGraphMapper.importGraph(TFGraphMapper.java:512)
at org.nd4j.imports.graphmapper.tf.TFGraphMapper.importGraph(TFGraphMapper.java:140)
at org.nd4j.imports.graphmapper.tf.TFGraphMapper.importGraph(TFGraphMapper.java:86)
at org.deeplearning4j.modelimportexamples.tf.advanced.bert.BertInferenceExample3.main(BertInferenceExample3.java:64)

thanks so much.

[agibsonccc]

@Maple that looks like a disconnected node found during import. Make sure the node has inputs and that every node has a constant or placeholder associated with it in the graph.

[Maple]

the nodes has inputs, and the model can be fine tuned correctly. i tried another way.for example:

i choose this node,“loss/Softmax”, in BertInferenceExample, it choose this too.and i can froze the graph to pb, like this:

but i run it in BertInferenceExample, it goes Exception.