Error converting my model trained with TF 2.1rc1 to TFLite #35449

aboerzel · 2019-12-27T13:27:25Z

System information

OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Windows 10
TensorFlow installed from (source or binary): conda
TensorFlow version (or github SHA if from source): 2.1.0-rc1

Provide the text output from tflite_convert

(dl4cv) D:\development\tensorflow\ANPR\output\rmsprop>tflite_convert --keras_model_file=clpr-model.h5 --enable_select_tf_ops --output_file=model.tflite
2019-12-27 14:09:27.630637: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.summary API due to missing TensorBoard installation.
2019-12-27 14:09:29.765186: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
2019-12-27 14:09:29.788663: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 2070 computeCapability: 7.5
coreClock: 1.83GHz coreCount: 36 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s
2019-12-27 14:09:29.797947: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2019-12-27 14:09:29.807131: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2019-12-27 14:09:29.815002: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2019-12-27 14:09:29.823010: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2019-12-27 14:09:29.830529: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2019-12-27 14:09:29.838455: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2019-12-27 14:09:29.850408: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2019-12-27 14:09:29.855640: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
2019-12-27 14:09:29.859642: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2019-12-27 14:09:29.865715: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 2070 computeCapability: 7.5
coreClock: 1.83GHz coreCount: 36 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s
2019-12-27 14:09:29.874362: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2019-12-27 14:09:29.878508: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2019-12-27 14:09:29.882420: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2019-12-27 14:09:29.886489: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2019-12-27 14:09:29.890738: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2019-12-27 14:09:29.894738: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2019-12-27 14:09:29.898962: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2019-12-27 14:09:29.903241: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
2019-12-27 14:09:30.584878: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-12-27 14:09:30.589782: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102]      0
2019-12-27 14:09:30.592385: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0:   N
2019-12-27 14:09:30.595864: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1241] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6271 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2070, pci bus id: 0000:01:00.0, compute capability: 7.5)
WARNING:tensorflow:No training configuration found in save file: the model was *not* compiled. Compile it manually.
W1227 14:09:33.236038 28348 hdf5_format.py:177] No training configuration found in save file: the model was *not* compiled. Compile it manually.
2019-12-27 14:09:33.443668: I tensorflow/core/grappler/devices.cc:55] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 1
2019-12-27 14:09:33.450829: I tensorflow/core/grappler/clusters/single_machine.cc:356] Starting new session
2019-12-27 14:09:33.467462: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 2070 computeCapability: 7.5
coreClock: 1.83GHz coreCount: 36 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s
2019-12-27 14:09:33.478633: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2019-12-27 14:09:33.483411: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2019-12-27 14:09:33.487932: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2019-12-27 14:09:33.492763: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2019-12-27 14:09:33.498057: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2019-12-27 14:09:33.502602: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2019-12-27 14:09:33.508437: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2019-12-27 14:09:33.513114: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
2019-12-27 14:09:33.516838: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-12-27 14:09:33.521102: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102]      0
2019-12-27 14:09:33.524027: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0:   N
2019-12-27 14:09:33.529383: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1241] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6271 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2070, pci bus id: 0000:01:00.0, compute capability: 7.5)
2019-12-27 14:09:33.597411: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: graph_to_optimize
2019-12-27 14:09:33.603519: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: Graph size after: 380 nodes (0), 484 edges (0), time = 17.869ms.
2019-12-27 14:09:33.610268: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: Graph size after: 380 nodes (0), 484 edges (0), time = 4.152ms.
2019-12-27 14:09:33.615867: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_1_backward_gru_1_while_cond_2105
2019-12-27 14:09:33.621215: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0.001ms.
2019-12-27 14:09:33.626682: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0.001ms.
2019-12-27 14:09:33.631716: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_forward_gru_while_body_1602
2019-12-27 14:09:33.637805: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0.001ms.
2019-12-27 14:09:33.643095: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0ms.
2019-12-27 14:09:33.647759: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_1_forward_gru_1_while_cond_1941
2019-12-27 14:09:33.653473: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0.001ms.
2019-12-27 14:09:33.658111: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0ms.
2019-12-27 14:09:33.663563: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_1_backward_gru_1_while_body_2106
2019-12-27 14:09:33.668843: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0.001ms.
2019-12-27 14:09:33.674032: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0ms.
2019-12-27 14:09:33.679432: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_backward_gru_while_body_1766
2019-12-27 14:09:33.685071: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0ms.
2019-12-27 14:09:33.690795: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0ms.
2019-12-27 14:09:33.695366: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_forward_gru_while_cond_1601
2019-12-27 14:09:33.701130: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0ms.
2019-12-27 14:09:33.705783: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0ms.
2019-12-27 14:09:33.710792: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_1_forward_gru_1_while_body_1942
2019-12-27 14:09:33.717048: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0.001ms.
2019-12-27 14:09:33.721844: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0ms.
2019-12-27 14:09:33.726940: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_backward_gru_while_cond_1765
2019-12-27 14:09:33.732340: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0ms.
2019-12-27 14:09:33.737188: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   function_optimizer: function_optimizer did nothing. time = 0ms.
2019-12-27 14:09:34.924139: I tensorflow/core/grappler/devices.cc:55] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 1
2019-12-27 14:09:34.931432: I tensorflow/core/grappler/clusters/single_machine.cc:356] Starting new session
2019-12-27 14:09:34.941771: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 2070 computeCapability: 7.5
coreClock: 1.83GHz coreCount: 36 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s
2019-12-27 14:09:34.952985: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2019-12-27 14:09:34.957078: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2019-12-27 14:09:34.962751: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2019-12-27 14:09:34.967762: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2019-12-27 14:09:34.973494: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2019-12-27 14:09:34.977944: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2019-12-27 14:09:34.983289: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2019-12-27 14:09:34.987946: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
2019-12-27 14:09:34.992126: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-12-27 14:09:34.996303: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102]      0
2019-12-27 14:09:34.999665: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0:   N
2019-12-27 14:09:35.004366: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1241] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6271 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2070, pci bus id: 0000:01:00.0, compute capability: 7.5)
2019-12-27 14:09:35.521865: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: graph_to_optimize
2019-12-27 14:09:35.526582: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 226 nodes (-16), 276 edges (-28), time = 227.782ms.
2019-12-27 14:09:35.532999: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 226 nodes (0), 276 edges (0), time = 111.153ms.
2019-12-27 14:09:35.538097: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_1_backward_gru_1_while_body_2106_frozen
2019-12-27 14:09:35.545175: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 60 nodes (-1), 79 edges (0), time = 2.335ms.
2019-12-27 14:09:35.550989: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 60 nodes (0), 79 edges (0), time = 0.737ms.
2019-12-27 14:09:35.556165: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_1_backward_gru_1_while_cond_2105_frozen
2019-12-27 14:09:35.562428: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 14 nodes (0), 4 edges (0), time = 0.306ms.
2019-12-27 14:09:35.567480: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 14 nodes (0), 4 edges (0), time = 0.176ms.
2019-12-27 14:09:35.573582: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_1_forward_gru_1_while_cond_1941_frozen
2019-12-27 14:09:35.579766: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 14 nodes (0), 4 edges (0), time = 0.262ms.
2019-12-27 14:09:35.584544: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 14 nodes (0), 4 edges (0), time = 0.175ms.
2019-12-27 14:09:35.590845: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_1_forward_gru_1_while_body_1942_frozen
2019-12-27 14:09:35.597525: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 60 nodes (-1), 79 edges (0), time = 1.396ms.
2019-12-27 14:09:35.602989: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 60 nodes (0), 79 edges (0), time = 0.698ms.
2019-12-27 14:09:35.609327: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_backward_gru_while_cond_1765_frozen
2019-12-27 14:09:35.614684: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 14 nodes (0), 4 edges (0), time = 0.274ms.
2019-12-27 14:09:35.621170: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 14 nodes (0), 4 edges (0), time = 0.216ms.
2019-12-27 14:09:35.626793: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_backward_gru_while_body_1766_frozen
2019-12-27 14:09:35.632917: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 60 nodes (-1), 79 edges (0), time = 1.417ms.
2019-12-27 14:09:35.639119: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 60 nodes (0), 79 edges (0), time = 0.711ms.
2019-12-27 14:09:35.645066: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_forward_gru_while_body_1602_frozen
2019-12-27 14:09:35.652559: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 60 nodes (-1), 79 edges (0), time = 1.349ms.
2019-12-27 14:09:35.658901: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 60 nodes (0), 79 edges (0), time = 0.749ms.
2019-12-27 14:09:35.665589: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:814] Optimization results for grappler item: model_bidirectional_forward_gru_while_cond_1601_frozen
2019-12-27 14:09:35.673058: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 14 nodes (0), 4 edges (0), time = 0.295ms.
2019-12-27 14:09:35.680300: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:816]   constant_folding: Graph size after: 14 nodes (0), 4 edges (0), time = 0.138ms.
Traceback (most recent call last):
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "C:\Users\Andreas\Anaconda3\envs\dl4cv\Scripts\tflite_convert.exe\__main__.py", line 7, in <module>
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\tensorflow_core\lite\python\tflite_convert.py", line 594, in main
    app.run(main=run_main, argv=sys.argv[:1])
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\tensorflow_core\python\platform\app.py", line 40, in run
    _run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\absl\app.py", line 299, in run
    _run_main(main, args)
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\absl\app.py", line 250, in _run_main
    sys.exit(main(argv))
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\tensorflow_core\lite\python\tflite_convert.py", line 577, in run_main
    _convert_tf2_model(tflite_flags)
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\tensorflow_core\lite\python\tflite_convert.py", line 235, in _convert_tf2_model
    tflite_model = converter.convert()
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\tensorflow_core\lite\python\lite.py", line 464, in convert
    **converter_kwargs)
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\tensorflow_core\lite\python\convert.py", line 457, in toco_convert_impl
    enable_mlir_converter=enable_mlir_converter)
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\tensorflow_core\lite\python\convert.py", line 203, in toco_convert_protos
    raise ConverterError("See console for info.\n%s\n%s\n" % (stdout, stderr))
tensorflow.lite.python.convert.ConverterError: See console for info.
2019-12-27 14:09:36.318110: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.summary API due to missing TensorBoard installation.
2019-12-27 14:09:39.097723: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2019-12-27 14:09:39.107524: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
2019-12-27 14:09:39.163607: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 2070 computeCapability: 7.5
coreClock: 1.83GHz coreCount: 36 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s
2019-12-27 14:09:39.164452: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2019-12-27 14:09:39.206090: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2019-12-27 14:09:39.249003: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2019-12-27 14:09:39.256358: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2019-12-27 14:09:39.307626: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2019-12-27 14:09:39.326748: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2019-12-27 14:09:39.414022: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2019-12-27 14:09:39.414730: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1697] Adding visible gpu devices: 0
2019-12-27 14:09:42.459316: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-12-27 14:09:42.459514: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102]      0
2019-12-27 14:09:42.459628: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0:   N
2019-12-27 14:09:42.460800: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1241] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 2903 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2070, pci bus id: 0000:01:00.0, compute capability: 7.5)
2019-12-27 14:09:42.536530: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListFromTensor
2019-12-27 14:09:42.536727: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.536928: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListFromTensor
2019-12-27 14:09:42.537124: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.537289: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListReserve
2019-12-27 14:09:42.537455: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.537616: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListReserve
2019-12-27 14:09:42.537804: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.537974: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: While
2019-12-27 14:09:42.538136: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.538282: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.538438: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: While
2019-12-27 14:09:42.538586: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.538734: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.538884: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListStack
2019-12-27 14:09:42.539041: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListStack
2019-12-27 14:09:42.539238: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListFromTensor
2019-12-27 14:09:42.539398: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.539579: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListFromTensor
2019-12-27 14:09:42.539737: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.539879: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListReserve
2019-12-27 14:09:42.540034: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.540195: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListReserve
2019-12-27 14:09:42.540357: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.540515: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: While
2019-12-27 14:09:42.540692: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.540861: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.541036: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: While
2019-12-27 14:09:42.541193: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.541338: I tensorflow/lite/toco/import_tensorflow.cc:193] Unsupported data type in placeholder op: 21
2019-12-27 14:09:42.541488: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListStack
2019-12-27 14:09:42.541642: I tensorflow/lite/toco/import_tensorflow.cc:659] Converting unsupported operation: TensorListStack
2019-12-27 14:09:42.550428: I tensorflow/lite/toco/graph_transformations/graph_transformations.cc:39] Before Removing unused ops: 98 operators, 264 arrays (0 quantized)
2019-12-27 14:09:42.555345: I tensorflow/lite/toco/graph_transformations/graph_transformations.cc:39] Before general graph transformations: 98 operators, 264 arrays (0 quantized)
2019-12-27 14:09:42.562862: I tensorflow/lite/toco/graph_transformations/graph_transformations.cc:39] After general graph transformations pass 1: 62 operators, 194 arrays (0 quantized)
2019-12-27 14:09:42.564023: I tensorflow/lite/toco/graph_transformations/graph_transformations.cc:39] After general graph transformations pass 2: 60 operators, 192 arrays (0 quantized)
2019-12-27 14:09:42.565092: I tensorflow/lite/toco/graph_transformations/graph_transformations.cc:39] After general graph transformations pass 3: 58 operators, 188 arrays (0 quantized)
2019-12-27 14:09:42.566092: I tensorflow/lite/toco/graph_transformations/graph_transformations.cc:39] Before Group bidirectional sequence lstm/rnn: 58 operators, 188 arrays (0 quantized)
2019-12-27 14:09:42.566900: I tensorflow/lite/toco/graph_transformations/graph_transformations.cc:39] Before dequantization graph transformations: 58 operators, 188 arrays (0 quantized)
2019-12-27 14:09:42.567646: I tensorflow/lite/toco/graph_transformations/graph_transformations.cc:39] Before Identify nearest upsample.: 58 operators, 188 arrays (0 quantized)
2019-12-27 14:09:42.569453: I tensorflow/lite/toco/allocate_transient_arrays.cc:345] Total transient array allocated size: 655360 bytes, theoretical optimal value: 655360 bytes.
2019-12-27 14:09:42.569852: I tensorflow/lite/toco/toco_tooling.cc:471] Number of parameters: 4121949
2019-12-27 14:09:42.575404: E tensorflow/lite/toco/toco_tooling.cc:498] We are continually in the process of adding support to TensorFlow Lite for more ops. It would be helpful if you could inform us of how this conversion went by opening a github issue at https://github.com/tensorflow/tensorflow/issues/new?template=40-tflite-op-request.md
 and pasting the following:

Some of the operators in the model are not supported by the standard TensorFlow Lite runtime. If those are native TensorFlow operators, you might be able to use the extended runtime by passing --enable_select_tf_ops, or by setting target_ops=TFLITE_BUILTINS,SELECT_TF_OPS when calling tf.lite.TFLiteConverter(). Otherwise, if you have a custom implementation for them you can disable this error with --allow_custom_ops, or by setting allow_custom_ops=True when calling tf.lite.TFLiteConverter(). Here is a list of builtin operators you are using: CONCATENATION, CONV_2D, DEPTHWISE_CONV_2D, FILL, FULLY_CONNECTED, MAX_POOL_2D, PACK, RESHAPE, REVERSE_V2, SHAPE, SOFTMAX, STRIDED_SLICE, TRANSPOSE. Here is a list of operators for which you will need custom implementations: TensorListFromTensor, TensorListReserve, TensorListStack, While.
Traceback (most recent call last):
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "C:\Users\Andreas\Anaconda3\envs\dl4cv\Scripts\toco_from_protos.exe\__main__.py", line 7, in <module>
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\tensorflow_core\lite\toco\python\toco_from_protos.py", line 93, in main
    app.run(main=execute, argv=[sys.argv[0]] + unparsed)
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\tensorflow_core\python\platform\app.py", line 40, in run
    _run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\absl\app.py", line 299, in run
    _run_main(main, args)
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\absl\app.py", line 250, in _run_main
    sys.exit(main(argv))
  File "c:\users\andreas\anaconda3\envs\dl4cv\lib\site-packages\tensorflow_core\lite\toco\python\toco_from_protos.py", line 56, in execute
    enable_mlir_converter)
Exception: We are continually in the process of adding support to TensorFlow Lite for more ops. It would be helpful if you could inform us of how this conversion went by opening a github issue at https://github.com/tensorflow/tensorflow/issues/new?template=40-tflite-op-request.md
 and pasting the following:

Some of the operators in the model are not supported by the standard TensorFlow Lite runtime. If those are native TensorFlow operators, you might be able to use the extended runtime by passing --enable_select_tf_ops, or by setting target_ops=TFLITE_BUILTINS,SELECT_TF_OPS when calling tf.lite.TFLiteConverter(). Otherwise, if you have a custom implementation for them you can disable this error with --allow_custom_ops, or by setting allow_custom_ops=True when calling tf.lite.TFLiteConverter(). Here is a list of builtin operators you are using: CONCATENATION, CONV_2D, DEPTHWISE_CONV_2D, FILL, FULLY_CONNECTED, MAX_POOL_2D, PACK, RESHAPE, REVERSE_V2, SHAPE, SOFTMAX, STRIDED_SLICE, TRANSPOSE. Here is a list of operators for which you will need custom implementations: TensorListFromTensor, TensorListReserve, TensorListStack, While.

Also, please include a link to a GraphDef or the model if possible.

download model

Any other info / logs

Include any logs or source code that would be helpful to diagnose the problem. If including tracebacks, please include the full traceback. Large logs and files should be attached.

GitHub project

Model architecture:

class OCR:
    @staticmethod
    def build(input_size, pool_size, output_size):
        conv_filters = 16
        kernel_size = (3, 3)
        time_dense_size = 32
        rnn_size = 512

        input_data = Input(name="input", shape=input_size)

        cnn = Conv2D(conv_filters, kernel_size, padding='same', kernel_initializer='he_normal')(input_data)
        cnn = BatchNormalization()(cnn)
        cnn = Activation('relu')(cnn)
        cnn = MaxPooling2D(pool_size=(pool_size, pool_size))(cnn)

        cnn = Conv2D(conv_filters, kernel_size, padding='same', kernel_initializer='he_normal')(cnn)
        cnn = BatchNormalization()(cnn)
        cnn = Activation('relu')(cnn)
        cnn = MaxPooling2D(pool_size=(pool_size, pool_size))(cnn)

        # CNN to RNN
        shape = cnn.get_shape()
        bgru = Reshape((shape[1], shape[2] * shape[3]))(cnn)

        bgru = Bidirectional(GRU(units=rnn_size, return_sequences=True, dropout=0.5))(bgru)
        bgru = TimeDistributed(Dense(units=time_dense_size))(bgru)

        bgru = Bidirectional(GRU(units=rnn_size, return_sequences=True, dropout=0.5))(bgru)
        output_data = TimeDistributed(Dense(units=output_size, activation="softmax"))(bgru)

        return input_data, output_data

The text was updated successfully, but these errors were encountered:

ymodak · 2020-01-02T19:39:04Z

The new experimental converter flag helps to convert models containing control flow ops in tf lite.
See https://www.tensorflow.org/api_docs/python/tf/lite/TFLiteConverter?version=nightly
Can you try with tf-nightly version and use experimental converter flag?

# `experimental_new_converter` flag.
converter.experimental_new_converter = True
tflite_model = converter.convert()

aboerzel · 2020-01-03T11:05:56Z

Thanks for the quick help, the problem seems to be solved in the tf-nightly version. I could now convert my Keras model to a TFLite model, see console output:


(tf-n) D:\development\tensorflow\saved-model>python
Python 3.7.3 (default, Mar 27 2019, 17:13:21) [MSC v.1915 64 bit (AMD64)] :: Anaconda, Inc. on win32

>>> import tensorflow as tf
2020-01-03 11:53:31.382855: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
WARNING:tensorflow:Falling back to tensorflow client, its recommended to install the cloud tpu client directly with pip install cloud-tpu-client .
>>> print (tf.__version__)
2.1.0-dev20200101
>>> from tensorflow_core.lite.python.lite import TFLiteConverter
>>> converter = TFLiteConverter.from_keras_model_file("clpr-model.h5")
2020-01-03 11:54:13.567519: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
2020-01-03 11:54:13.597923: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1558] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 2070 computeCapability: 7.5
coreClock: 1.83GHz coreCount: 36 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s
2020-01-03 11:54:13.608151: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2020-01-03 11:54:13.617085: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2020-01-03 11:54:13.624166: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2020-01-03 11:54:13.631583: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2020-01-03 11:54:13.641627: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2020-01-03 11:54:13.654835: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2020-01-03 11:54:13.697939: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2020-01-03 11:54:13.705541: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1700] Adding visible gpu devices: 0
2020-01-03 11:54:13.711667: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2020-01-03 11:54:13.718275: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1558] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 2070 computeCapability: 7.5
coreClock: 1.83GHz coreCount: 36 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s
2020-01-03 11:54:13.727735: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2020-01-03 11:54:13.732743: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2020-01-03 11:54:13.737100: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2020-01-03 11:54:13.741393: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2020-01-03 11:54:13.745770: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2020-01-03 11:54:13.750174: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2020-01-03 11:54:13.754456: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2020-01-03 11:54:13.758860: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1700] Adding visible gpu devices: 0
2020-01-03 11:54:16.071893: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1099] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-01-03 11:54:16.078369: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1105]      0
2020-01-03 11:54:16.082532: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1118] 0:   N
2020-01-03 11:54:16.087348: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1244] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6304 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2070, pci bus id: 0000:01:00.0, compute capability: 7.5)
WARNING:tensorflow:No training configuration found in save file: the model was *not* compiled. Compile it manually.
2020-01-03 11:54:22.032670: I tensorflow/core/grappler/devices.cc:55] Number of eligible GPUs (core count >= 8, compute capability >= 0.0): 1
2020-01-03 11:54:22.041300: I tensorflow/core/grappler/clusters/single_machine.cc:356] Starting new session
2020-01-03 11:54:22.073419: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1558] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 2070 computeCapability: 7.5
coreClock: 1.83GHz coreCount: 36 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.29GiB/s
2020-01-03 11:54:22.086076: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_101.dll
2020-01-03 11:54:22.091592: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll
2020-01-03 11:54:22.096409: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll
2020-01-03 11:54:22.100644: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll
2020-01-03 11:54:22.106354: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll
2020-01-03 11:54:22.111275: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll
2020-01-03 11:54:22.115978: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2020-01-03 11:54:22.122779: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1700] Adding visible gpu devices: 0
2020-01-03 11:54:22.126064: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1099] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-01-03 11:54:22.131080: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1105]      0
2020-01-03 11:54:22.133956: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1118] 0:   N
2020-01-03 11:54:22.142183: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1244] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6304 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2070, pci bus id: 0000:01:00.0, compute capability: 7.5)
2020-01-03 11:54:22.292642: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:815] Optimization results for grappler item: graph_to_optimize
2020-01-03 11:54:22.299483: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: Graph size after: 380 nodes (0), 484 edges (0), time = 23.395ms.
2020-01-03 11:54:22.305784: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: Graph size after: 380 nodes (0), 484 edges (0), time = 4.831ms.
2020-01-03 11:54:22.311102: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:815] Optimization results for grappler item: model_bidirectional_1_forward_gru_1_while_cond_1537
2020-01-03 11:54:22.317628: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 2.427ms.
2020-01-03 11:54:22.322884: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
2020-01-03 11:54:22.328516: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:815] Optimization results for grappler item: model_bidirectional_forward_gru_while_cond_1197
2020-01-03 11:54:22.334748: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
2020-01-03 11:54:22.340884: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
2020-01-03 11:54:22.345895: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:815] Optimization results for grappler item: model_bidirectional_forward_gru_while_body_1198
2020-01-03 11:54:22.351964: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0.001ms.
2020-01-03 11:54:22.356766: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
2020-01-03 11:54:22.362160: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:815] Optimization results for grappler item: model_bidirectional_backward_gru_while_body_1362
2020-01-03 11:54:22.368032: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
2020-01-03 11:54:22.372883: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
2020-01-03 11:54:22.377991: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:815] Optimization results for grappler item: model_bidirectional_1_backward_gru_1_while_cond_1701
2020-01-03 11:54:22.383364: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
2020-01-03 11:54:22.388489: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
2020-01-03 11:54:22.393606: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:815] Optimization results for grappler item: model_bidirectional_1_backward_gru_1_while_body_1702
2020-01-03 11:54:22.398977: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
2020-01-03 11:54:22.404466: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0.001ms.
2020-01-03 11:54:22.409196: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:815] Optimization results for grappler item: model_bidirectional_backward_gru_while_cond_1361
2020-01-03 11:54:22.415716: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0.001ms.
2020-01-03 11:54:22.421434: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
2020-01-03 11:54:22.427168: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:815] Optimization results for grappler item: model_bidirectional_1_forward_gru_1_while_body_1538
2020-01-03 11:54:22.433035: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0.001ms.
2020-01-03 11:54:22.437814: I tensorflow/core/grappler/optimizers/meta_optimizer.cc:817]   function_optimizer: function_optimizer did nothing. time = 0ms.
>>> converter.experimental_new_converter = True
>>> tflite_model = converter.convert()
>>> open("clpr-model.tflite", "wb").write(tflite_model)
16612140
>>>

tensorflow-bot · 2020-01-03T11:05:58Z

Are you satisfied with the resolution of your issue?
Yes
No

geetachavan1 added the TF 2.1 for tracking issues in 2.1 release label Dec 27, 2019

oanush self-assigned this Dec 30, 2019

oanush added comp:lite TF Lite related issues type:bug Bug labels Dec 30, 2019

oanush assigned ymodak and unassigned oanush Dec 30, 2019

ymodak added the stat:awaiting response Status - Awaiting response from author label Jan 2, 2020

aboerzel closed this as completed Jan 3, 2020

gadagashwini-zz mentioned this issue Jan 31, 2020

Some of the operators in the model are not supported by the standard TensorFlow Lite runtime. #36015

Closed

This was referenced Feb 20, 2020

Custom implementations for AvgPool3D, Conv3D, MaxPool3D #36895

Closed

Here is a list of operators for which you will need custom implementations: RandomUniform. #37009

Closed

This was referenced Mar 5, 2020

Does BatchMatMul is still not supported in TF1.15? #37247

Closed

CombinedNonMaxSuppression is not whitelisted operator #37301

Closed

amahendrakar mentioned this issue Mar 16, 2020

need custom implementations: NonMaxSuppressionV5 #37608

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Error converting my model trained with TF 2.1rc1 to TFLite #35449

Error converting my model trained with TF 2.1rc1 to TFLite #35449

aboerzel commented Dec 27, 2019

ymodak commented Jan 2, 2020

aboerzel commented Jan 3, 2020

tensorflow-bot bot commented Jan 3, 2020

Error converting my model trained with TF 2.1rc1 to TFLite #35449

Error converting my model trained with TF 2.1rc1 to TFLite #35449

Comments

aboerzel commented Dec 27, 2019

ymodak commented Jan 2, 2020

aboerzel commented Jan 3, 2020

tensorflow-bot bot commented Jan 3, 2020