MO error from tensorflow model

[Ryan_YP_H_]

Hi,
I'm trying transform tensorflow model to OpenVINO IR files.

There are 2 kinds of models.

1. Tensorflow deeplab Model zoo (deeplab)
2. Fast Style Transfer

First model is from tensorflow/models, I follow OpenVINO steps to convert model by below cmd line: 
python /opt/intel/computer_vision_sdk/deployment_tools/model_optimizer/mo_tf.py \
> --input_model frozen_inference_graph.pb \
> --input_shape [1,513,513,3]

And I got ERROR message:

Model Optimizer arguments:
Common parameters:
        - Path to the Input Model:      /home/Ryan/realtime_segmenation/models/deeplabv3_mnv2_pascal_train_aug/frozen_inference_graph.pb
        - Path for generated IR:        /home/Ryan/realtime_segmenation/models/deeplabv3_mnv2_pascal_train_aug/.
        - IR output name:       frozen_inference_graph
        - Log level:    ERROR
        - Batch:        Not specified, inherited from the model
        - Input layers:         Not specified, inherited from the model
        - Output layers:        Not specified, inherited from the model
        - Input shapes:         [1,513,513,3]
        - Mean values:  Not specified
        - Scale values:         Not specified
        - Scale factor:         Not specified
        - Precision of IR:      FP32
        - Enable fusing:        True
        - Enable grouped convolutions fusing:   True
        - Move mean values to preprocess section:       False
        - Reverse input channels:       False
TensorFlow specific parameters:
        - Input model in text protobuf format:  False
        - Offload unsupported operations:       False
        - Path to model dump for TensorBoard:   None
        - Update the configuration file with input/output node names:   None
        - Use configuration file used to generate the model with Object Detection API:  None
        - Operations to offload:        None
        - Patterns to offload:  None
        - Use the config file:  None
Model Optimizer version:        1.2.185.5335e231
/opt/intel/computer_vision_sdk_fpga_2018.3.343/deployment_tools/model_optimizer/mo/front/common/partial_infer/slice.py:90: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.
  value = value[slice_idx]
[ ERROR ]  Cannot infer shapes or values for node "GreaterEqual".
[ ERROR ]  Input 0 of node GreaterEqual was passed int64 from add_1_port_0_ie_placeholder:0 incompatible with expected int32.
[ ERROR ]
[ ERROR ]  It can happen due to bug in custom shape infer function <function tf_native_tf_node_infer at 0x7f415f23c620>.
[ ERROR ]  Or because the node inputs have incorrect values/shapes.
[ ERROR ]  Or because input shapes are incorrect (embedded to the model or passed via --input_shape).
[ ERROR ]  Run Model Optimizer with --log_level=DEBUG for more information.
[ ERROR ]  Stopped shape/value propagation at "GreaterEqual" node.
 For more information please refer to Model Optimizer FAQ (<INSTALL_DIR>/deployment_tools/documentation/docs/MO_FAQ.html), question #38.

------------------
Second model is a style transfer model, graph build in python code.
You can download the pretrain model from the github provide link.

The pretrain model are metagraph file, and I got error when convert metagraph > IR
python3.6 /opt/intel/computer_vision_sdk/deployment_tools/model_optimizer/mo_tf.py \
> --input_meta_graph udnie.ckpt.meta

Model Optimizer arguments:
Common parameters:
        - Path to the Input Model:      None
        - Path for generated IR:        /home/Ryan/realtime_segmenation/computex/style_model/.
        - IR output name:       udnie.ckpt
        - Log level:    ERROR
        - Batch:        Not specified, inherited from the model
        - Input layers:         Not specified, inherited from the model
        - Output layers:        Not specified, inherited from the model
        - Input shapes:         Not specified, inherited from the model
        - Mean values:  Not specified
        - Scale values:         Not specified
        - Scale factor:         Not specified
        - Precision of IR:      FP32
        - Enable fusing:        True
        - Enable grouped convolutions fusing:   True
        - Move mean values to preprocess section:       False
        - Reverse input channels:       False
TensorFlow specific parameters:
        - Input model in text protobuf format:  False
        - Offload unsupported operations:       False
        - Path to model dump for TensorBoard:   None
        - Update the configuration file with input/output node names:   None
        - Use configuration file used to generate the model with Object Detection API:  None
        - Operations to offload:        None
        - Patterns to offload:  None
        - Use the config file:  None
Model Optimizer version:        1.2.185.5335e231
[ ERROR ]  Cannot load input model: Unsuccessful TensorSliceReader constructor: Failed to find any matching files for /home/Ryan/realtime_segmenation/computex/style_model/udnie.ckp
         [[Node: save/RestoreV2_81 = RestoreV2[dtypes=[DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2_81/tensor_names, save/RestoreV2_81/shape_and_slices)]]

Caused by op 'save/RestoreV2_81', defined at:
  File "/opt/intel/computer_vision_sdk/deployment_tools/model_optimizer/mo_tf.py", line 31, in <module>
    sys.exit(main(get_tf_cli_parser(), 'tf'))
  File "/opt/intel/computer_vision_sdk_fpga_2018.3.343/deployment_tools/model_optimizer/mo/main.py", line 321, in main
    return driver(argv)
  File "/opt/intel/computer_vision_sdk_fpga_2018.3.343/deployment_tools/model_optimizer/mo/main.py", line 263, in driver
    mean_scale_values=mean_scale)
  File "/opt/intel/computer_vision_sdk_fpga_2018.3.343/deployment_tools/model_optimizer/mo/pipeline/tf.py", line 80, in tf2nx
    saved_model_tags=argv.saved_model_tags)
  File "/opt/intel/computer_vision_sdk_fpga_2018.3.343/deployment_tools/model_optimizer/mo/front/tf/loader.py", line 140, in load_tf_graph_def
    restorer = tf.train.import_meta_graph(input_meta_graph_def)
  File "/usr/lib/python3.6/site-packages/tensorflow/python/training/saver.py", line 1960, in import_meta_graph
    **kwargs)
  File "/usr/lib/python3.6/site-packages/tensorflow/python/framework/meta_graph.py", line 744, in import_scoped_meta_graph
    producer_op_list=producer_op_list)
  File "/usr/lib/python3.6/site-packages/tensorflow/python/util/deprecation.py", line 432, in new_func
    return func(*args, **kwargs)
  File "/usr/lib/python3.6/site-packages/tensorflow/python/framework/importer.py", line 442, in import_graph_def
    _ProcessNewOps(graph)
  File "/usr/lib/python3.6/site-packages/tensorflow/python/framework/importer.py", line 234, in _ProcessNewOps
    for new_op in graph._add_new_tf_operations(compute_devices=False):  # pylint: disable=protected-access
  File "/usr/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3563, in _add_new_tf_operations
    for c_op in c_api_util.new_tf_operations(self)
  File "/usr/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3563, in <listcomp>
    for c_op in c_api_util.new_tf_operations(self)
  File "/usr/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3450, in _create_op_from_tf_operation
    ret = Operation(c_op, self)
  File "/usr/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1740, in __init__
    self._traceback = self._graph._extract_stack()  # pylint: disable=protected-access

NotFoundError (see above for traceback): Unsuccessful TensorSliceReader constructor: Failed to find any matching files for /home/Ryan/realtime_segmenation/computex/style_model/udnie.ckp
         [[Node: save/RestoreV2_81 = RestoreV2[dtypes=[DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2_81/tensor_names, save/RestoreV2_81/shape_and_slices)]]

I try to freeze this model in python code after sess.run

frozen = tf.graph_util.convert_variables_to_constants(sess, sess.graph_def, ["clip_by_value"])
graph_io.write_graph(frozen, './style_model/IR/', 'inference_graph.pb', as_text=False)

And convert this .pb file by MO, still got error:

Model Optimizer arguments:
Common parameters:
        - Path to the Input Model:      /home/Ryan/realtime_segmenation/computex/style_model/IR/inference_graph.pb
        - Path for generated IR:        /home/Ryan/realtime_segmenation/computex/style_model/IR/.
        - IR output name:       inference_graph
        - Log level:    ERROR
        - Batch:        Not specified, inherited from the model
        - Input layers:         Not specified, inherited from the model
        - Output layers:        Not specified, inherited from the model
        - Input shapes:         Not specified, inherited from the model
        - Mean values:  Not specified
        - Scale values:         Not specified
        - Scale factor:         Not specified
        - Precision of IR:      FP32
        - Enable fusing:        True
        - Enable grouped convolutions fusing:   True
        - Move mean values to preprocess section:       False
        - Reverse input channels:       False
TensorFlow specific parameters:
        - Input model in text protobuf format:  False
        - Offload unsupported operations:       False
        - Path to model dump for TensorBoard:   None
        - Update the configuration file with input/output node names:   None
        - Use configuration file used to generate the model with Object Detection API:  None
        - Operations to offload:        None
        - Patterns to offload:  None
        - Use the config file:  None
Model Optimizer version:        1.2.185.5335e231
[ ERROR ]  Shape is not defined for output 0 of "Slice".
[ ERROR ]  Cannot infer shapes or values for node "Slice".
[ ERROR ]  Not all output shapes were inferred or fully defined for node "Slice".
 For more information please refer to Model Optimizer FAQ (<INSTALL_DIR>/deployment_tools/documentation/docs/MO_FAQ.html), question #40.
[ ERROR ]
[ ERROR ]  It can happen due to bug in custom shape infer function <function tf_slice_infer at 0x7f028badb730>.
[ ERROR ]  Or because the node inputs have incorrect values/shapes.
[ ERROR ]  Or because input shapes are incorrect (embedded to the model or passed via --input_shape).
[ ERROR ]  Run Model Optimizer with --log_level=DEBUG for more information.
[ ERROR ]  Stopped shape/value propagation at "Slice" node.
 For more information please refer to Model Optimizer FAQ (<INSTALL_DIR>/deployment_tools/documentation/docs/MO_FAQ.html), question #38.

My question is

1. I try to refrozen first model by export_model.py in the github project, got error in same. How can I fix the error ?
2. The style transfer model need to build graph and restore variable from meta, so I try to save the graph from python code after sess.run, but it got an error about Slice layer. How to fix the error ?

Thanks in advanced.
Best regard,

Ryan.

[Shubha_R_Intel · Accepted solution]

Ryan for the TensorFlow DeepLab model please try the following command (it works for me):

python mo.py --scale 1 --model_name test_deeplabv3_mobilenet_v2_argmax_tfFP32CPU1Truesync --input_shape "(1,513,513,3)" --input 1:mul_1 --input_model  "c:\Users\sdramani\Downloads\deeplabv3_mnv2_pascal_train_aug\frozen_inference_graph.pb" --framework tf --output_dir c:\Users\sdramani\Downloads\out_dir --data_type FP32 --output ArgMax

In the output_dir you should see the following 3 files created:

test_deeplabv3_mobilenet_v2_argmax_tfFP32CPU1Truesync.xml

test_deeplabv3_mobilenet_v2_argmax_tfFP32CPU1Truesync.mapping

test_deeplabv3_mobilenet_v2_argmax_tfFP32CPU1Truesync.bin

As for Fast Style Transfer we are still investigating this. Stay tuned.

Thanks for using OpenVino !

Shubha

[Ryan_YP_H_]

Shubha R. (Intel) wrote:

Ryan for the TensorFlow DeepLab model please try the following command (it works for me):

python mo.py --scale 1 --model_name test_deeplabv3_mobilenet_v2_argmax_tfFP32CPU1Truesync --input_shape "(1,513,513,3)" --input 1:mul_1 --input_model  "c:\Users\sdramani\Downloads\deeplabv3_mnv2_pascal_train_aug\frozen_inference_graph.pb" --framework tf --output_dir c:\Users\sdramani\Downloads\out_dir --data_type FP32 --output ArgMax

In the output_dir you should see the following 3 files created:

test_deeplabv3_mobilenet_v2_argmax_tfFP32CPU1Truesync.xml

test_deeplabv3_mobilenet_v2_argmax_tfFP32CPU1Truesync.mapping

test_deeplabv3_mobilenet_v2_argmax_tfFP32CPU1Truesync.bin

As for Fast Style Transfer we are still investigating this. Stay tuned.

Thanks for using OpenVino !

Shubha

Thanks Shubha,

These IR files work fine in my project.

So it needs to define model input and output layer when got MO error ?