OpenVino Model Optimizer-Shape [-1 -1 -1 1] is not fully defined for output 0 of "image"

[rachel_r_Intel]

I am trying to help convert trained tensorflow model hfnet, I have no idea about more information about this model(I don't know the --input_shape parameter).this is the folder structure.

.
├── hfnet
│   ├── checkpoint
│   ├── config.yaml
│   ├── events.out.tfevents.1541966416.lo-s4-033
│   ├── log
│   ├── model.ckpt-83096.data-00000-of-00001
│   └── model.ckpt-83096.index
└── saved_models
    └── hfnet
        ├── saved_model.pb
        └── variables
            ├── variables.data-00000-of-00001
            └── variables.index

command:

python3 mo_tf.py --saved_model_dir /home/intel/hfnet/saved_models/hfnet/

I get the following error:

Common parameters:
	- Path to the Input Model: 	None
	- Path for generated IR: 	/opt/intel/openvino_2019.3.376/deployment_tools/model_optimizer/.
	- IR output name: 	saved_model
	- 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
	- 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
	- Operations to offload: 	None
	- Patterns to offload: 	None
	- Use the config file: 	None
Model Optimizer version: 	2019.3.0-408-gac8584cb7
[ ERROR ]  Shape [-1 -1 -1  1] is not fully defined for output 0 of "image". Use --input_shape with positive integers to override model input shapes.
[ ERROR ]  Cannot infer shapes or values for node "image".
[ ERROR ]  Not all output shapes were inferred or fully defined for node "image". 
 For more information please refer to Model Optimizer FAQ (https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_Model_Optimizer_FAQ.html), question #40. 
[ ERROR ]  
[ ERROR ]  It can happen due to bug in custom shape infer function <function Parameter.__init__.<locals>.<lambda> at 0x7fc05e765598>.
[ 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 ]  Not all output shapes were inferred or fully defined for node "image". 
 For more information please refer to Model Optimizer FAQ (https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_Model_Optimizer_FAQ.html), question #40. 
Stopped shape/value propagation at "image" node. 
 For more information please refer to Model Optimizer FAQ (https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_Model_Optimizer_FAQ.html), question #38. 
Exception occurred during running replacer "REPLACEMENT_ID" (<class 'extensions.middle.PartialInfer.PartialInfer'>): Not all output shapes were inferred or fully defined for node "image". 
 For more information please refer to Model Optimizer FAQ (https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_Model_Optimizer_FAQ.html), question #40. 
Stopped shape/value propagation at "image" node. 
 For more information please refer to Model Optimizer FAQ (https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_Model_Optimizer_FAQ.html), question #38. 

Any help?Thanks

[Monique_J_Intel]

Hi Rachel,

You can analyze your model using summarize graph tool and that should be able to tell you information about your inputs and outputs. The python tool is in /opt/intel/openvino/deployment_tools/model_optimizer/mo/utils/. To understand how to run it type summarize_graph.py -h.

Also taking a look at the github and knowing Tensorflow extensively your input tensor is in the form of [Batch,Height,Width,Channels] with color channels being 1.

So if you know that you'll only be processing 1 image at a time and you know the height and width of the test images you can fill this in for example  if your image resolution is 300x300 then try 

python3 mo_tf.py --saved_model_dir /home/intel/hfnet/saved_models/hfnet/ --input_shape [1,300,300,1] 

Let me know if this helps.

Kind Regards,

Monique Jones

[daisyyang]

Hi Monique,

Thanks for sharing your idea. I got the same error as Rachel, but I'm trying to convert an onnx model. Looks like summarize_graph tool is only for TensorFlow model? Do you know is there any tool can get the input shape of an onnx model?

Thanks,

Daisy