Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.
6403 Discussions

Exception occurred during running replacer "REPLACEMENT_ID"

HA14
Beginner
623 Views

Hello, I'm trying to convert a .pb Tensorflow (1.14v) with @

but I'm keep getting this error with input shape:

[ ERROR ] Cannot infer shapes or values for node "replica_0_1/lambda_7/Slice".
[ ERROR ] Output shape: [ 0 768 512 3] of node "replica_0_1/lambda_7/Slice" contains non-positive values
[ ERROR ]
[ ERROR ] It can happen due to bug in custom shape infer function <function Slice.infer at 0x000001F24BFBB3A0>.
[ 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).

This is the command I'm using "python mo.py --input_model <Input_Model>  --input inputs_mask,inputs_img --input_shape [1,768,512,3],[1,768,512,3]" 

I also tried [1,3, 768,512],[1,3,768,512] and tried the -b 1 solution, I got the same error with all of them.

Only when I tried -b 1 I got the a different shape in the error above 

[ ERROR ] Cannot infer shapes or values for node "replica_2_1/lambda_11/Slice".
[ ERROR ] Output shape: [ 0 512 768 3] of node "replica_2_1/lambda_11/Slice" contains non-positive values
[ ERROR ]
[ ERROR ] It can happen due to bug in custom shape infer function <function Slice.infer at 0x000002577C4A13A0>.
[ 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).

Labels (1)
0 Kudos
2 Replies
Adli
Moderator
611 Views

Hi HA14,


Thank you for reaching out to us. If possible, could you share additional information regarding the Tensorflow model:

  • Is it a custom model or a public model?
  • If it is a custom model, please specify which topology is applied to the model.
  • If it is a public model, please specify which model.


If it is a custom model, please verify the model layers/operations are supported by the OpenVINO toolkit. To see the layers supported by the Tensorflow framework, please refer to the following link:

https://docs.openvinotoolkit.org/latest/openvino_docs_MO_DG_prepare_model_Supported_Frameworks_Layers.html#tensorflow_supported_operations


Regards,

Adli


0 Kudos
Adli
Moderator
588 Views

Hi HA14,


Thank you for your question. If you need any additional information from Intel, please submit a new question as this thread is no longer being monitored.


Regards,

Adli


0 Kudos
Reply