Viktor Gyenes
Viktor Gyenes
# Ask a Question ### Question Is it possible to export ONNX models with tensor shapes included so that shape inference need not be run when importing a model? How...
# Ask a Question ### Question I have seen some (experimental?) support in ONNX proto to functions (sequence of ops). Is it actually supported by ONNX latest version, or was...
I have been investigating the ONNX version of SSD models (SSD Resnet, SSD Mobilenet) by opening them in Netron viewer. The ONNX models contain a few Loop operators; more specifically,...
I believe I have found a FW bug in the implementation of the output subcommand 'GotoAbsolutePosition' for _virtual_ ports (both commands 0x0D and 0x0E) when using _immediate_ execution that happens...
As discussed in another topic (#1476), a program is not really able to tell if the contents of the persistent storage are valid for the given program or not, because...
**Describe the bug** In many cases there are `If` operators in a model that compare shapes. When these shapes are known, the condition becomes constant and the whole `If` operator...
**Describe the bug** I am trying to run ONNX simplifier on a model from MLPerf (retinanet-resnext50 800x800). However, the simplifier crashes with a segmentation fault, seems like a null pointer...
**Describe the bug** When simplifying a model, it can happen that the model gets much larger than it was originally and as a consequence exceeds the maximum size of 2GB...
The C API for the NNEF parser (currently nnef_parser.h and .c in this project) has been merged into the NNEF-Tools project (using the name cnnef.h and .c), which is referenced...