When exporting a model with python3 model_inspect.py --runmode=saved_model --tflite_path=model.tflite, the exported TensorFlow Lite model is incompatible with mobile libraries like MediaPipe and the TensorFlow Lite Task Library, since they require the model to have four output tensors (as described here), while the exported model is an EfficientDetNet model with only one output tensor.
Hi!
When exporting a model with
python3 model_inspect.py --runmode=saved_model --tflite_path=model.tflite
, the exported TensorFlow Lite model is incompatible with mobile libraries like MediaPipe and the TensorFlow Lite Task Library, since they require the model to have four output tensors (as described here), while the exported model is anEfficientDetNet
model with only one output tensor.If I understand correctly, the full model required by those libraries is
EfficientDetModel
, added in commit https://github.com/google/automl/commit/cedf982590e1d7da0e388a8ae01d3e0d6c3a8625.Would it be possible to improve the
model_inspect.py
script to export the TFLite model with all the four required tensors?Thanks!