Hello,
I am new in SNPE SDK and I am trying to convert a model writen in Keras 2.30-tf / Tensorflow 2.2 from .pb format into .dlc
using snpe-tensorflow-to-dlc. The .dlc file generation completes with the following WARNING :
WARNING_TF_OP_NOT_SUPPORTED: Operation (dummy_convnet/convT1/Shape) of type (Shape) is not supported by converter.
I get the following error with the dummy model here below which has a Conv2DTranspose Layer that seems to cause the issue.
The SNPE SDK documentation specifies that conv2d_transpose is supported, is there an issue with TF2.2 Conv2DTranspose Layer ?
>>>>>>>>>>
>>>>>>>>>>
I did not get any reply so far but I've made some progress but still facing some issues.
The dummy model containing a Conv2D Layer was written in Keras/TF2.2.0 and converted to a frozen model in .pb format.
Then I used SNPE v 1.51.0 using TF1.15 to convert it to a .dlc file , and that produced the WARNING_TF_OP_NOT_SUPPORTED mentioned above
In order to avoid compatibility issues , I used TF2.2.0 also to do the .pb => .dlc conversion (same version as used to generate the .pb).
Now I have an error during the .pb => .dlc conversion:
>>>>>>>>>>>
>>>>>>>>>>>
It tells me that the Tensor at the input of my Conv2D Layer should be of int32 type, but it should be a float32, How can it be int32, all neural networks manipulate floats and matmul operations. What do you think is the issue here ?.
Just to summary my configurations :
Keras to generate (and train) the model : 2.3.0-tf
tensorflow for converted .keras model file to .pb : 2.2.0
To convert .pb to .dlc:
ubuntu-18 LTS , SNPE: snpe-1.51.0
Tried 2 options :
- tensorflow: 1.15.0 and python 3.6.9 (Got the WARNING_TF_OP_NOT_SUPPORTED)
- tensorflow: 2.2.0 and python 3.6.9 (Got the ValueError: Tensor conversion requested dtype int32 for Tensor with dtype float32)
Just to summary my configurations :
Keras to generate (and train) the model : 2.3.0-tf
tensorflow for converted .keras model file to .pb : 2.2.0
To convert .pb to .dlc:
ubuntu-18 LTS , SNPE: snpe-1.51.0
Tried 2 options :
- tensorflow: 1.15.0 and python 3.6.9 (Got the WARNING_TF_OP_NOT_SUPPORTED)
- tensorflow: 2.2.0 and python 3.6.9 (Got the ValueError: Tensor conversion requested dtype int32 for Tensor with dtype float32)