Community
cancel
Showing results for 
Search instead for 
Did you mean: 
195 Views

Issue in optimizing text classifier BERT model built using Tensorflow

(base) dhanu@myVM:~$ python3 /opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo_tf.py --input_model /home/dhanu/Downloads/model4_256/output/graph.pbtxt --input "IteratorGetNext:0[16 256],IteratorGetNext:1[16 256],IteratorGetNext:2[16],IteratorGetNext:3[16 10],IteratorGetNext:4[16 256]" --input_model_is_text --disable_nhwc_to_nchw --keep_shape_ops
Model Optimizer arguments:
Common parameters:
    - Path to the Input Model:     /home/dhanu/Downloads/model4_256/output/graph.pbtxt
    - Path for generated IR:     /home/dhanu/.
    - IR output name:     graph
    - Log level:     ERROR
    - Batch:     Not specified, inherited from the model
    - Input layers:     IteratorGetNext:0[16 256],IteratorGetNext:1[16 256],IteratorGetNext:2[16],IteratorGetNext:3[16 10],IteratorGetNext:4[16 256]
    - 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:     True
    - Path to model dump for TensorBoard:     None
    - List of shared libraries with TensorFlow custom layers implementation:     None
    - Update the configuration file with input/output node names:     None
    - Use configuration file used to generate the model with Object Detection API:     None
    - Use the config file:     None
Model Optimizer version:     2020.2.0-60-g0bc66e26ff
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
[ ERROR ]  -------------------------------------------------
[ ERROR ]  ----------------- INTERNAL ERROR ----------------
[ ERROR ]  Unexpected exception happened.
[ ERROR ]  Please contact Model Optimizer developers and forward the following information:
[ ERROR ]  Exception occurred during running replacer "None (<class 'extensions.front.no_op_eraser.NoOpEraser'>)": The node group_deps_1 must have just one input
[ ERROR ]  Traceback -(most recent call last):
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/utils/class_registration.py", line 288, in apply_transform
    for_graph_and_each_sub_graph_recursively(graph, replacer.find_and_replace_pattern)
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/middle/pattern_match.py", line 58, in for_graph_and_each_sub_graph_recursively
    func(graph)
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/front/common/replacement.py", line 150, in find_and_replace_pattern
    apply_pattern(graph, action=self.replace_sub_graph, **self.pattern())
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/middle/pattern_match.py", line 95, in apply_pattern
    action(graph, match)
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/extensions/front/no_op_eraser.py", line 40, in replace_sub_graph
    graph.erase_node(match['noop'])
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/graph/graph.py", line 626, in erase_node
    assert len(inputs) <= 1, "The node {} must have just one input".format(node.soft_get('name'))
AssertionError: The node group_deps_1 must have just one input

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/main.py", line 307, in main
    return driver(argv)
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/main.py", line 272, in driver
    ret_res = emit_ir(prepare_ir(argv), argv)
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/main.py", line 237, in prepare_ir
    graph = unified_pipeline(argv)
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/pipeline/unified.py", line 29, in unified_pipeline
    class_registration.ClassType.BACK_REPLACER
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/utils/class_registration.py", line 334, in apply_replacements
    apply_replacements_list(graph, replacers_order)
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/utils/class_registration.py", line 324, in apply_replacements_list
    num_transforms=len(replacers_order))
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/utils/logger.py", line 124, in wrapper
    function(*args, **kwargs)
  File "/opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo/utils/class_registration.py", line 312, in apply_transform
    )) from err
Exception: Exception occurred during running replacer "None (<class 'extensions.front.no_op_eraser.NoOpEraser'>)": The node group_deps_1 must have just one input

[ ERROR ]  ---------------- END OF BUG REPORT --------------
[ ERROR ]  -------------------------------------------------

 

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Also tried with the following command

(base) dhanu@myVM:~$ python3 /opt/intel/openvino_2020.2.120/deployment_tools/model_optimizer/mo_tf.py --input_model /home/dhanu/Downloads/model4_256/output/graph.pbtxt --input_checkpoint /home/dhanu/Downloads/model4_256/output/model.ckpt-3000.data-00000-of-00001 --input_model_is_text
Model Optimizer arguments:
Common parameters:
    - Path to the Input Model:     /home/dhanu/Downloads/model4_256/output/graph.pbtxt
    - Path for generated IR:     /home/dhanu/.
    - IR output name:     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:     True
    - Path to model dump for TensorBoard:     None
    - List of shared libraries with TensorFlow custom layers implementation:     None
    - Update the configuration file with input/output node names:     None
    - Use configuration file used to generate the model with Object Detection API:     None
    - Use the config file:     None
Model Optimizer version:     2020.2.0-60-g0bc66e26ff
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:516: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:517: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:518: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:519: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:520: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorflow/python/framework/dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:541: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:542: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:543: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:544: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:545: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
/home/dhanu/anaconda3/lib/python3.7/site-packages/tensorboard/compat/tensorflow_stub/dtypes.py:550: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
[ FRAMEWORK ERROR ]  Cannot load input model: Op type not registered 'MapAndBatchDataset' in binary running on myVM. Make sure the Op and Kernel are registered in the binary running in this process. Note that if you are loading a saved graph which used ops from tf.contrib, accessing (e.g.) `tf.contrib.resampler` should be done before importing the graph, as contrib ops are lazily registered when the module is first accessed.

 

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Note: The model is fine tuned using BERT uncased_L-12_H-768_A-12 with tensorflow 1.14

Tensorflow version - 1.14

Openvino version - openvino_2020.2.120

OS - Ubuntu

CPU specs:

Architecture:        x86_64
CPU op-mode(s):      32-bit, 64-bit
Byte Order:          Little Endian
CPU(s):              2
On-line CPU(s) list: 0,1
Thread(s) per core:  2
Core(s) per socket:  1
Socket(s):           1
NUMA node(s):        1
Vendor ID:           GenuineIntel
CPU family:          6
Model:               85
Model name:          Intel(R) Xeon(R) Platinum 8171M CPU @ 2.60GHz
Stepping:            4
CPU MHz:             2095.193
BogoMIPS:            4190.38
Virtualization:      VT-x
Hypervisor vendor:   Microsoft
Virtualization type: full
L1d cache:           32K
L1i cache:           32K
L2 cache:            1024K
L3 cache:            36608K
NUMA node0 CPU(s):   0,1
Flags:               fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology cpuid pni pclmulqdq vmx ssse3 fma cx16 pcid sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti tpr_shadow vnmi ept vpid fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f avx512dq rdseed adx smap clflushopt avx512cd avx512bw avx512vl xsaveopt xsavec xsaves

0 Kudos
1 Reply
Munesh_Intel
Moderator
195 Views

Hi Dhanunjay,

I’ve replicated your issue using a supported pre-trained BERT model.

I would suggest you use ‘inference_graph.pb’ input model instead.

Detailed information is available in the section “Convert Reshape-able TensorFlow BERT Model to the Intermediate Representation”, at the following page:

https://docs.openvinotoolkit.org/2020.3/_docs_MO_DG_prepare_model_convert_model_tf_specific_Convert_...

 

Regards,

Munesh

Reply