Hi Val,

bai__val · ‎09-12-2019

Hello!

I'm working in Ubuntu and using Python API

I'm trying to run my YOLO3 Mobilenet model in Neural Compute Stick.

When I run: sudo python3 mo.py --input_model /home/Desktop/keras_to_tensorflow-master/logs/tf_model.pb --input_shape [1,416,416,3],[1,13,13,3,10],[1,26,26,3,10],[1,52,52,3,10] --input input_1,input_2,input_3,input_4 --output yolo_loss/Select, I get this error

[ 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 "REPLACEMENT_ID (<class 'extensions.front.input_cut.InputCut'>)": Attempt to access node input_3 that not in graph
[ ERROR ] Traceback (most recent call last):
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/utils/class_registration.py", line 271, in apply_replacements
replacer.find_and_replace_pattern(graph)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/extensions/front/input_cut.py", line 34, in find_and_replace_pattern
add_input_ops(graph, graph.graph['user_shapes'], True)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/front/extractor.py", line 1118, in add_input_ops
smart_node = Node(graph, node_id)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/graph/graph.py", line 35, in __init__
assert node in graph, "Attempt to access node {} that not in graph".format(node)
AssertionError: Attempt to access node input_3 that not in graph

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

Traceback (most recent call last):
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/main.py", line 302, in main
return driver(argv)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/main.py", line 251, in driver
is_binary=not argv.input_model_is_text)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/pipeline/tf.py", line 133, in tf2nx
class_registration.apply_replacements(graph, class_registration.ClassType.FRONT_REPLACER)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/utils/class_registration.py", line 299, in apply_replacements
)) from err
Exception: Exception occurred during running replacer "REPLACEMENT_ID (<class 'extensions.front.input_cut.InputCut'>)": Attempt to access node input_3 that not in graph

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

Sahira_Intel · ‎09-16-2019

Hi Val,

I see this is a converted model from Keras to Tensorflow - did you convert yourself? Can you please attach your model so I can take a look at it?

Best Regards,

Sahira

bai__val · ‎09-18-2019

Hello, Sahira.

Thank you for your comment. It is a private model , would you please PM me and I can privately send you the model.

I converted model from Keras to Tensorflow using keras_to_tensorflow.py https://github.com/amir-abdi/keras_to_tensorflow

When I run:

sudo python3 mo_tf.py --input_model /home/Desktop/keras_to_tensorflow-master/logs/tf_model.pb --input_shape [1,416,416,3] --input input_1 --output conv2d_7/BiasAdd,conv2d_15/BiasAdd,conv2d_23/BiasAdd

i get this error:

[ ERROR ] List of operations that cannot be converted to Inference Engine IR:
[ ERROR ] AddV2 (47)
[ ERROR ] conv1_bn/batchnorm/add_1
[ ERROR ] conv_dw_1_bn/batchnorm/add_1
[ ERROR ] conv_pw_1_bn/batchnorm/add_1
[ ERROR ] conv_dw_2_bn/batchnorm/add_1
[ ERROR ] conv_pw_2_bn/batchnorm/add_1
[ ERROR ] conv_dw_3_bn/batchnorm/add_1
[ ERROR ] conv_pw_3_bn/batchnorm/add_1
[ ERROR ] conv_dw_4_bn/batchnorm/add_1
[ ERROR ] conv_pw_4_bn/batchnorm/add_1
[ ERROR ] conv_dw_5_bn/batchnorm/add_1
[ ERROR ] conv_pw_5_bn/batchnorm/add_1
[ ERROR ] conv_dw_6_bn/batchnorm/add_1
[ ERROR ] conv_pw_6_bn/batchnorm/add_1
[ ERROR ] conv_dw_7_bn/batchnorm/add_1
[ ERROR ] conv_pw_7_bn/batchnorm/add_1
[ ERROR ] conv_dw_8_bn/batchnorm/add_1
[ ERROR ] conv_pw_8_bn/batchnorm/add_1
[ ERROR ] conv_dw_9_bn/batchnorm/add_1
[ ERROR ] conv_pw_9_bn/batchnorm/add_1
[ ERROR ] conv_dw_10_bn/batchnorm/add_1
[ ERROR ] conv_pw_10_bn/batchnorm/add_1
[ ERROR ] conv_dw_11_bn/batchnorm/add_1
[ ERROR ] conv_pw_11_bn/batchnorm/add_1
[ ERROR ] conv_dw_12_bn/batchnorm/add_1
[ ERROR ] conv_pw_12_bn/batchnorm/add_1
[ ERROR ] conv_dw_13_bn/batchnorm/add_1
[ ERROR ] conv_pw_13_bn/batchnorm/add_1
[ ERROR ] batch_normalization_1/batchnorm/add_1
[ ERROR ] batch_normalization_2/batchnorm/add_1
[ ERROR ] batch_normalization_3/batchnorm/add_1
[ ERROR ] batch_normalization_4/batchnorm/add_1
[ ERROR ] batch_normalization_5/batchnorm/add_1
[ ERROR ] batch_normalization_7/batchnorm/add_1
[ ERROR ] batch_normalization_8/batchnorm/add_1
[ ERROR ] batch_normalization_9/batchnorm/add_1
[ ERROR ] batch_normalization_10/batchnorm/add_1
[ ERROR ] batch_normalization_11/batchnorm/add_1
[ ERROR ] batch_normalization_12/batchnorm/add_1
[ ERROR ] batch_normalization_13/batchnorm/add_1
[ ERROR ] batch_normalization_14/batchnorm/add_1
[ ERROR ] batch_normalization_15/batchnorm/add_1
[ ERROR ] batch_normalization_16/batchnorm/add_1
[ ERROR ] batch_normalization_17/batchnorm/add_1
[ ERROR ] batch_normalization_18/batchnorm/add_1
[ ERROR ] batch_normalization_19/batchnorm/add_1
[ ERROR ] batch_normalization_20/batchnorm/add_1
[ ERROR ] batch_normalization_6/batchnorm/add_1

then I tried to run:

sudo python3 mo_tf.py --input_model /home/Desktop/keras_to_tensorflow-master/logs/tf_model.pb --input_shape [1,416,416,3] --input input_1 --output conv2d_7/BiasAdd,conv2d_15/BiasAdd,conv2d_23/BiasAdd --tensorflow_operation_patterns AddV2

I get this error

[ 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 "REPLACEMENT_ID (<class 'extensions.middle.CustomSubgraphCall.CustomSubgraphCall'>)": Attempt to connect conv1_bn/moving_variance/read/Output_0/Data_ to Reshape_. Node Reshape_ has no in port (0)
[ ERROR ] Traceback (most recent call last):
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/utils/class_registration.py", line 273, in apply_replacements
for_graph_and_each_sub_graph_recursively(graph, replacer.find_and_replace_pattern)
File "/opt/intel/openvino_2019.2.275/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_2019.2.275/deployment_tools/model_optimizer/extensions/middle/CustomSubgraphCall.py", line 322, in find_and_replace_pattern
CustomSubgraphCall.add_reshapes_for_tf_subgraph_calls(graph)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/extensions/middle/CustomSubgraphCall.py", line 262, in add_reshapes_for_tf_subgraph_calls
CustomSubgraphCall.add_reshape_before_op_node(graph, src_node_name, dst_node_name, edge_attrs)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/extensions/middle/CustomSubgraphCall.py", line 182, in add_reshape_before_op_node
(reshaped_data_node_name, op_node_name, edge_attrs)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/graph/graph.py", line 546, in add_edges_from
self.add_edge(u, v, key=key, **ddd)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/graph/graph.py", line 523, in add_edge
assert vnode.has_port('in', attr['in']), "{} Node {} has no in port ({})".format(message, vnode.name, attr['in'])
AssertionError: Attempt to connect conv1_bn/moving_variance/read/Output_0/Data_ to Reshape_. Node Reshape_ has no in port (0)

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

Traceback (most recent call last):
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/main.py", line 302, in main
return driver(argv)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/main.py", line 251, in driver
is_binary=not argv.input_model_is_text)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/pipeline/tf.py", line 134, in tf2nx
class_registration.apply_replacements(graph, class_registration.ClassType.MIDDLE_REPLACER)
File "/opt/intel/openvino_2019.2.275/deployment_tools/model_optimizer/mo/utils/class_registration.py", line 299, in apply_replacements
)) from err
Exception: Exception occurred during running replacer "REPLACEMENT_ID (<class 'extensions.middle.CustomSubgraphCall.CustomSubgraphCall'>)": Attempt to connect conv1_bn/moving_variance/read/Output_0/Data_ to Reshape_. Node Reshape_ has no in port (0)

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

Sahira_Intel · ‎09-20-2019

Hi Val,

I have sent you a PM. Please respond there.

Best Regards,

Sahira

Unexpected exception happened. Please contact Model Optimizer developers and forward the following information