[GNA] Added fix multiple output with one go to memory and test
[GNA] Added fix multiple output with one go to memory and test
[GNA] Added fix multiple output with one go to memory and test
Added multi output
Update gna_pass_manager.cpp
test
[GNA] Added fix multiple output with one go to memory and test
[GNA] Added fix multiple output with one go to memory and test
[GNA] Added fix multiple output with one go to memory and test
Added multi output
Update gna_pass_manager.cpp
test
tests
[GNA] Added fix multiple output with one go to memory and test
[GNA] Added fix multiple output with one go to memory and test
Added multi output
Update gna_pass_manager.cpp
test
tests
Added pass
Test
test
tests_2
return old
The problem was in order of freeing memory. _context was removed before
_device and it looks like cl::Device in destructor tries to read some
info from cl::Context. And in this case we got this problem with
addressing because the memory already was freed.
For fixing the problem I changed the order of constructing members. And
based on principle: "First constructed, last destructed", the problem
was fixed.
JIRA: 29649
* Fix kaldi models (batch > 1)
* ngraph codestyle
* fix ngraph to ie conversion
* Added comment
* apply review comments
* Added test for the case using the SetBatchSize function when ReadValue op is in the network
* Check status code instead of message
* Use new ngraph api
* Removed back phase transformations related to IRv7
* Fixed setting value for the input port using the 'set_value' method
* Removed front and middle phase transformations related to IRv7
* Cleanup the rest of the Model Optimizer transformations from IRv7 specific transformations
* Final cleanup of the deprecated IR v7 related code
* Removed 'blobs_as_input' usage in the Model Optimizer.
* Removed function '_fuse_add' from the Model Optimizer since it is not used anymore.
* Removed 'keep_in_IR' node attribute for FakeQuantize ops in the MO
* Disabled failing gpu_engine.user_context test
* Fix build issue
Why:
* Enable to build OpenVINO.
This change addresses the need by:
* Adding include directories,
* Removing IE::inference_engine_c_api dependency.
* Remove IE::inference_engine_nn_builder reference.
Why:
* Enable to build OpenVINO.
This change addresses the need by:
* Removing IE::inference_engine_nn_builder dependency.
Co-authored-by: Ilya Lavrenov <ilya.lavrenov@intel.com>
For b_fs_yx_fsv16 format in reference kernel features for dispatch are
rounded to multiple of 16. This change adds correct check in kernel to
return work-items that are inside this dispatch padding.
Previously those work-items could corrupt memory expected to be filled
with 0s, and for parametrized activation due to bounds checking with
modulo operator they could have been corrupting actual layer output.
Issue: CVS-27672
