keras_to_tensorflow
keras_to_tensorflow copied to clipboard
Associate MetaGraphDef with tag Serve
Hello,
I'm trying to convert Pb model to TFLite with following code:
converter = tf.contrib.lite.TocoConverter.from_saved_model(DIR_OUT + '/test')
tflite_model = converter.convert()
open("converted_model.tflite", "wb").write(tflite_model)
but there's an error:
RuntimeError: MetaGraphDef associated with tags {'serve'} could not be found in SavedModel. To inspect available tag-sets in the SavedModel, please use the SavedModel CLI: saved_model_cli
When used, SavedModel CLI show empty result, so my question is, how to add it to saved Pb properly?
The same problem occured to me.Have you solved it?Looking forward to your reply!
If I remember correctly, it wasn't solved.
Project required iOS mlmodel anyway, so I converted pb to it, and reduced size ( https://developer.apple.com/documentation/coreml/reducing_the_size_of_your_core_ml_app )
Same problem here. If someone has an idea of how to solve this please share
any solution to this ?