Exporting tflite model for tensorflow lite TPU

[zlg9folira]

Has a TFLite model ever been exported for the TF model in this repo? If there is one, can you upload it or a step-by-step guide is appreciated.

I run TF2.1.0 Python3.7 Linux and am trying to export one using
python3 export_tflite_ssd_graph.py --pipeline_config_path=/PATH_TO/pipeline.config --trained_checkpoint_prefix=/PATH_TO/model.ckpt.data-00000-of-00001 --output_directory=/my_tflite --add_postprocessing_op=true

and get error
File "../.local/lib/python3.7/site-packages/tensorflow_core/python/ops/array_ops.py", line 2715, in placeholder raise RuntimeError("tf.placeholder() is not compatible with " RuntimeError: tf.placeholder() is not compatible with eager execution.

[dtsmith2001]

@zlg9folira we are having a similar problem

File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/lite/python/tflite_convert.py", line 235, in _convert_tf2_model tflite_model = converter.convert() File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/lite/python/lite.py", line 446, in convert "invalid shape '{1}'.".format(_get_tensor_name(tensor), shape_list)) ValueError: None is only supported in the 1st dimension. Tensor 'image_tensor' has invalid shape '[None, None, None, 3]'.

Did you resolve the problem or develop a work-around?