search

emedvedev / attention-ocr

A Tensorflow model for text recognition (CNN + seq2seq with visual attention) available as a Python package and compatible with Google Cloud ML Engine.
MIT License
1.08k stars 256 forks source link

convert exported model to tensorflow lite? #136

Open Luukjn opened 5 years ago

Luukjn commented 5 years ago

Hello,

I'm trying to run the exported model on a mobile device. After some research I found Tensorflow Lite. Is there a way to convert the exported model to Tensorflow Lite? If there is, how would I go about doing just that?

Luukjn commented 5 years ago

I have some more information on this issue now. I ran the following command:

 tflite_convert --output_file=converted_model.tflite --graph_def_file=frozen_graph.pb --input_arrays=input_image_as_bytes --input_shapes=1,200,150,3 --output_arrays=prediction,probability

and I got the following output:

2019-06-14 15:42:35.173030: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2019-06-14 15:42:35.893291: I tensorflow/core/grappler/devices.cc:60] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 0 (Note: TensorFlow was not compiled with CUDA support)
2019-06-14 15:42:35.900314: I tensorflow/core/grappler/clusters/single_machine.cc:359] Starting new session
2019-06-14 15:42:37.480768: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:716] Optimization results for grappler item: graph_to_optimize
2019-06-14 15:42:37.485294: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:718]   constant folding: Graph size after: 2388 nodes (-101), 3426 edges (-116), time = 1084.72705ms.
2019-06-14 15:42:37.490429: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:718]   constant folding: Graph size after: 2388 nodes (0), 3426 edges (0), time = 348.687ms.
Traceback (most recent call last):
  File "C:\Program Files\Python36\Lib\runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "C:\Program Files\Python36\Lib\runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "C:\environments\Scripts\tflite_convert.exe\__main__.py", line 9, in <module>
  File "c:\environments\lib\site-packages\tensorflow\lite\python\tflite_convert.py", line 503, in main
    app.run(main=run_main, argv=sys.argv[:1])
  File "c:\environments\lib\site-packages\tensorflow\python\platform\app.py", line 40, in run
    _run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
  File "c:\environments\lib\site-packages\absl\app.py", line 300, in run
    _run_main(main, args)
  File "c:\environments\lib\site-packages\absl\app.py", line 251, in _run_main
    sys.exit(main(argv))
  File "c:\environments\lib\site-packages\tensorflow\lite\python\tflite_convert.py", line 499, in run_main
    _convert_tf1_model(tflite_flags)
  File "c:\environments\lib\site-packages\tensorflow\lite\python\tflite_convert.py", line 193, in _convert_tf1_model
    output_data = converter.convert()
  File "c:\environments\lib\site-packages\tensorflow\lite\python\lite.py", line 912, in convert
    **converter_kwargs)
  File "c:\environments\lib\site-packages\tensorflow\lite\python\convert.py", line 404, in toco_convert_impl
    input_data.SerializeToString())
  File "c:\environments\lib\site-packages\tensorflow\lite\python\convert.py", line 172, in toco_convert_protos
    "TOCO failed. See console for info.\n%s\n%s\n" % (stdout, stderr))
tensorflow.lite.python.convert.ConverterError: TOCO failed. See console for info.
2019-06-14 15:42:39.712616: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: MutableHashTableV2
2019-06-14 15:42:39.713274: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "NoOp" device_type: "CPU"') for unknown op: NoOp
2019-06-14 15:42:39.713579: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "NoOp" device_type: "GPU"') for unknown op: NoOp
2019-06-14 15:42:39.713705: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "_HostRecv" device_type: "GPU" host_memory_arg: "tensor"') for unknown op: _HostRecv
2019-06-14 15:42:39.714031: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "_Send" device_type: "CPU"') for unknown op: _Send
2019-06-14 15:42:39.714445: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "_HostRecv" device_type: "CPU"') for unknown op: _HostRecv
2019-06-14 15:42:39.714550: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "_Send" device_type: "GPU"') for unknown op: _Send
2019-06-14 15:42:39.714645: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "_Recv" device_type: "CPU"') for unknown op: _Recv
2019-06-14 15:42:39.714775: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "_HostSend" device_type: "GPU" host_memory_arg: "tensor"') for unknown op: _HostSend
2019-06-14 15:42:39.714942: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "_Recv" device_type: "GPU"') for unknown op: _Recv
2019-06-14 15:42:39.715539: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "_HostSend" device_type: "CPU"') for unknown op: _HostSend
2019-06-14 15:42:39.715906: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "WrapDatasetVariant" device_type: "CPU"') for unknown op: WrapDatasetVariant
2019-06-14 15:42:39.716129: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "WrapDatasetVariant" device_type: "GPU" host_memory_arg: "input_handle" host_memory_arg: "output_handle"') for unknown op: WrapDatasetVariant
2019-06-14 15:42:39.716583: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "UnwrapDatasetVariant" device_type: "CPU"') for unknown op: UnwrapDatasetVariant
2019-06-14 15:42:39.716781: E tensorflow/core/framework/op_kernel.cc:1501] OpKernel ('op: "UnwrapDatasetVariant" device_type: "GPU" host_memory_arg: "input_handle" host_memory_arg: "output_handle"') for unknown op: UnwrapDatasetVariant
2019-06-14 15:42:39.717264: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: MutableHashTableV2
2019-06-14 15:42:39.717523: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.717784: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.718041: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.718185: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.718438: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.718616: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.718772: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.718950: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.719122: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: LookupTableInsertV2
2019-06-14 15:42:39.719460: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.719583: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.719664: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.719875: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.720138: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.720308: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.720615: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.720776: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.721034: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.721207: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.721474: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.721643: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.721825: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.721984: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.722171: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.722331: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.722661: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArrayV3
2019-06-14 15:42:39.722879: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: TensorArrayV3
2019-06-14 15:42:39.723015: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArrayV3
2019-06-14 15:42:39.723234: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: TensorArrayV3
2019-06-14 15:42:39.723484: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.723578: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.723664: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.723864: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.723952: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.724032: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.724110: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.724432: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.724571: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArrayScatterV3
2019-06-14 15:42:39.724668: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: TensorArrayScatterV3
2019-06-14 15:42:39.724929: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-06-14 15:42:39.725159: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Enter
2019-06-14 15:42:39.725439: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: LoopCond
2019-06-14 15:42:39.725610: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: LoopCond
2019-06-14 15:42:39.725729: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Exit
2019-06-14 15:42:39.725892: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: Exit
2019-06-14 15:42:39.726182: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArrayReadV3
2019-06-14 15:42:39.726343: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: TensorArrayReadV3
2019-06-14 15:42:39.726544: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArraySizeV3
2019-06-14 15:42:39.726742: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: TensorArraySizeV3
2019-06-14 15:42:39.726938: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: DecodePng
2019-06-14 15:42:39.727105: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: DecodePng
2019-06-14 15:42:39.727308: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArrayGatherV3
2019-06-14 15:42:39.727563: I tensorflow/lite/toco/import_tensorflow.cc:1385] Unable to determine output type for op: TensorArrayGatherV3
2019-06-14 15:42:39.727820: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 2
2019-06-14 15:42:39.727981: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 2
2019-06-14 15:42:39.728341: F tensorflow/lite/toco/import_tensorflow.cc:1557] Check failed: data_type == DT_FLOAT
Fatal Python error: Aborted

Current thread 0x000019ec (most recent call first):
  File "c:\environments\lib\site-packages\tensorflow\lite\toco\python\toco_from_protos.py", line 33 in execute
  File "c:\environments\lib\site-packages\absl\app.py", line 251 in _run_main
  File "c:\environments\lib\site-packages\absl\app.py", line 300 in run
  File "c:\environments\lib\site-packages\tensorflow\python\platform\app.py", line 40 in run
  File "c:\environments\lib\site-packages\tensorflow\lite\toco\python\toco_from_protos.py", line 59 in main
  File "C:\environments\Scripts\toco_from_protos.exe\__main__.py", line 9 in <module>
  File "C:\Program Files\Python36\Lib\runpy.py", line 85 in _run_code
  File "C:\Program Files\Python36\Lib\runpy.py", line 193 in _run_module_as_main

I'm trying to run the conversion on Windows 10 using tf-nightly.

emedvedev commented 5 years ago

The Python API for the converter doesn't work with frozen graphs from TF 1.0, and I guess it might be the case with tflite_convert, too. Try exporting as SavedModel instead of a frozen graph, and then run tflite_convert on the generated directory.

You could also expand the export function yourself and use the tf.lite.TFLiteConverter Python API—please do submit a PR then, it would be much appreciated. :)

Luukjn commented 5 years ago

@emedvedev Alas I can't use the saved_model since my requirments are that the model must run on the new Edge TPU. And the TPU does not support dynamic sizes. (https://coral.withgoogle.com/docs/edgetpu/models-intro/) I don't really know what would be required to create a model that meets the requirements for the TPU.

emedvedev commented 5 years ago

Not entirely sure how a SavedModel would be different from the frozen graph of the same model for conversion purposes, so can’t advise there.

If you’re sure that SavedModel is out, it seems like your best bet would be extending the code in the export method and use the Python API for conversion. On Jun 17, 2019, 17:19 +0700, Luukjn notifications@github.com, wrote:

@emedvedev Alas I can't use the saved_model since my requirments are that the model must run on the new Edge TPU. And the TPU does not support dynamic sizes. (https://coral.withgoogle.com/docs/edgetpu/models-intro/) I don't really know what would be required to create a model that meets the requirements for the TPU. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

emedvedev commented 5 years ago

Sorry for the delay, but I think I've got it. I used MutableHashTable for the charmap lookup, and it's not supported by TF Lite. However, a simple HashTable can be used instead, so if you change L159-L169 in model.py to:

table = tf.HashTable(
    tf.KeyValueTensorInitializer(
        tf.constant(list(range(len(DataGen.CHARMAP))), dtype=tf.int64), 
        tf.constant(DataGen.CHARMAP),
    ), 
    -1,
)

And remove the with tf.control_dependencies statement on L171, then it should work. It might make the model init somewhat faster, too, although not by much.

chbaumgaertner commented 5 years ago

I'm also trying to convert the model to tf.lite and encounter the same problems. After changing L159-L169 and L171, I get the following error:

Traceback (most recent call last):
  File "/home/chris/tf/tf/bin/aocr", line 10, in <module>
    sys.exit(main())
  File "/home/chris/tf/tf/lib/python3.6/site-packages/aocr/__main__.py", line 252, in main
    channels=parameters.channels,
  File "/home/chris/tf/tf/lib/python3.6/site-packages/aocr/model/model.py", line 166, in __init__
    -1
  File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/contrib/lookup/lookup_ops.py", line 332, in __init__
    super(HashTable, self).__init__(default_value, initializer)
  File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/ops/lookup_ops.py", line 167, in __init__
    default_value, dtype=self._value_dtype)
  File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1087, in convert_to_tensor
    return convert_to_tensor_v2(value, dtype, preferred_dtype, name)
  File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1145, in convert_to_tensor_v2
    as_ref=False)
  File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1224, in internal_convert_to_tensor
    ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
  File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 305, in _constant_tensor_conversion_function
    return constant(v, dtype=dtype, name=name)
  File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 246, in constant
    allow_broadcast=True)
  File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 284, in _constant_impl
    allow_broadcast=allow_broadcast))
  File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py", line 466, in make_tensor_proto
    _AssertCompatible(values, dtype)
  File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py", line 371, in _AssertCompatible
    (dtype.name, repr(mismatch), type(mismatch).__name__))
TypeError: Expected string, got -1 of type 'int' instead.

Any recommendations?

emedvedev commented 5 years ago

Could be something with the HashTable syntax. My recommendation was off the top of my head, so not guaranteed, but should still be pretty close :)

Is it just the export that doesn’t work with the changes? Does training work? If it doesn’t, then it’s definitely the HashTable syntax, and should be straightforward to fix through some debugging and going through the docs. On Jul 9, 2019, 15:09 +0300, B.Sc. Christopher Baumgärtner notifications@github.com, wrote:

I'm also trying to convert the model to tf.lite and encounter the same problems. After changing L159-L169 and L171, I get the following error: Traceback (most recent call last): File "/home/chris/tf/tf/bin/aocr", line 10, in sys.exit(main()) File "/home/chris/tf/tf/lib/python3.6/site-packages/aocr/main.py", line 252, in main channels=parameters.channels, File "/home/chris/tf/tf/lib/python3.6/site-packages/aocr/model/model.py", line 166, in init -1 File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/contrib/lookup/lookup_ops.py", line 332, in init super(HashTable, self).init(default_value, initializer) File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/ops/lookup_ops.py", line 167, in init default_value, dtype=self._value_dtype) File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1087, in convert_to_tensor return convert_to_tensor_v2(value, dtype, preferred_dtype, name) File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1145, in convert_to_tensor_v2 as_ref=False) File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1224, in internal_convert_to_tensor ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref) File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 305, in _constant_tensor_conversion_function return constant(v, dtype=dtype, name=name) File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 246, in constant allow_broadcast=True) File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 284, in _constant_impl allow_broadcast=allow_broadcast)) File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py", line 466, in make_tensor_proto _AssertCompatible(values, dtype) File "/home/chris/tf/tf/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py", line 371, in _AssertCompatible (dtype.name, repr(mismatch), type(mismatch).name)) TypeError: Expected string, got -1 of type 'int' instead. Any recommendations? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

chrimarf commented 5 years ago

Hi there, I am trying to do the same on a Mac. I have been able to export in tflite mobilenet V1 but I am not able to do the same with this model even if I do it with the saved_model. do you have an update on the hashtable issue ? Actually by changing the lines below, the training doesn't work anymore.

My code

aocr train --max-prediction=4 --max-width=400 cropped_images/datasets/training.tfrecords

tflite_convert --output_file=converted_model.tflite --saved_model_dir=/tmp/saved_model --input_arrays=input_image_as_bytes --input_shapes=1,300,300,3 --output_arrays=prediction,probability

Then I get this error 

  File "/usr/local/bin/tflite_convert", line 10, in <module>
    sys.exit(main())
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/lite/python/tflite_convert.py", line 503, in main
    app.run(main=run_main, argv=sys.argv[:1])
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/platform/app.py", line 40, in run
    _run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
  File "/usr/local/lib/python2.7/dist-packages/absl/app.py", line 300, in run
    _run_main(main, args)
  File "/usr/local/lib/python2.7/dist-packages/absl/app.py", line 251, in _run_main
    sys.exit(main(argv))
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/lite/python/tflite_convert.py", line 499, in run_main
    _convert_tf1_model(tflite_flags)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/lite/python/tflite_convert.py", line 193, in _convert_tf1_model
    output_data = converter.convert()
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/lite/python/lite.py", line 898, in convert
    **converter_kwargs)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/lite/python/convert.py", line 404, in toco_convert_impl
    input_data.SerializeToString())
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/lite/python/convert.py", line 172, in toco_convert_protos
    "TOCO failed. See console for info.\n%s\n%s\n" % (stdout, stderr))
tensorflow.lite.python.convert.ConverterError: TOCO failed. See console for info.
2019-11-14 09:15:42.689165: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: MutableHashTableV2
2019-11-14 09:15:42.700047: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 20
2019-11-14 09:15:42.700221: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.700337: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.700447: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 2
2019-11-14 09:15:42.700490: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.700544: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.700581: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: LookupTableInsertV2
2019-11-14 09:15:42.700983: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701018: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 20
2019-11-14 09:15:42.701050: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701174: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701224: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701299: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701339: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 20
2019-11-14 09:15:42.701373: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701410: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701451: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701515: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArrayV3
2019-11-14 09:15:42.701566: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 20
2019-11-14 09:15:42.701601: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArrayV3
2019-11-14 09:15:42.701631: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 20
2019-11-14 09:15:42.701659: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701719: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701755: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 20
2019-11-14 09:15:42.701784: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701826: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701856: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 20
2019-11-14 09:15:42.701888: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArrayScatterV3
2019-11-14 09:15:42.701941: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Enter
2019-11-14 09:15:42.701972: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: LoopCond
2019-11-14 09:15:42.702028: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: Exit
2019-11-14 09:15:42.702138: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArrayReadV3
2019-11-14 09:15:42.702192: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArraySizeV3
2019-11-14 09:15:42.702898: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: DecodePng
2019-11-14 09:15:42.702986: I tensorflow/lite/toco/import_tensorflow.cc:1336] Converting unsupported operation: TensorArrayGatherV3
2019-11-14 09:15:42.703074: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 2
2019-11-14 09:15:42.703117: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 2
2019-11-14 09:15:42.734079: F tensorflow/lite/toco/import_tensorflow.cc:1542] Check failed: data_type == DT_FLOAT 
Aborted
OctaM commented 4 years ago

Any update?

tunymedia02 commented 4 years ago

Same problem here. Any update to this?