* Updated submodule onnx to 10.5
Signed-off-by: Michal Guzek <mguzek@nvidia.com>
* Updated files to 10.5
Signed-off-by: Michal Guzek <mguzek@nvidia.com>
* Restore the previous .gitmodules
Signed-off-by: Michal Guzek <mguzek@nvidia.com>
* .gitmodules new line
Signed-off-by: Michal Guzek <mguzek@nvidia.com>
* Update public docker files
Signed-off-by: Michal Guzek <mguzek@nvidia.com>
* Update 10.5 oss deleted files
Signed-off-by: Michal Guzek <mguzek@nvidia.com>
* Update deprecated APIs
Signed-off-by: Michal Guzek <mguzek@nvidia.com>
---------
Signed-off-by: Michal Guzek <mguzek@nvidia.com>
when TRT_LOCAL_SM is defined for develop build, we should only enable
sm-specifc source list, so that we can reduce the build time/library size.
I also removed the dependency on LEGACY_FUSEDEPSEP_ARCH and
AMPERE_FUSEDEPSEP_ARCH, they are deprecated.
Signed-off-by: Shuyue Lan <shuyuel@nvidia.com>
Signed-off-by: Rajeev Rao <rajeevrao@nvidia.com>
Clean up CustomFCPlugin
Clean up CropAndResizePlugin
Clean up BatchTilePlugin
Improve code quality of proposalPlugin
Clean up DetectionLayerPlugin
Run clang-format on all .cpp and .h plugin code
upper case literal suffixes in plugin module
Signed-off-by: Shuyue Lan <shuyuel@nvidia.com>
Signed-off-by: Rajeev Rao <rajeevrao@nvidia.com>
Also adds goldenIO testcases: one for class agnostic NMS and another for class aware NMS.
With this, golden_io_snippet.py is also modified to support checking for multiple
plugin outputs.
Adds `class_agnostic` parameter to `EfficientNMS_TRT` plugin:
- Some object detection networks/architectures like YOLO series need to
use class-independent NMS operations.
- If `class_agnostic` is enabled, class-independent NMS is performed;
otherwise, different classes would do NMS separately.
Signed-off-by: Samurdhi Karunaratne <skarunaratne@nvidia.com>
Signed-off-by: Shuyue Lan <shuyuel@nvidia.com>
Signed-off-by: Rajeev Rao <rajeevrao@nvidia.com>
use kINT32 type for mask output in embLayerNormVarSeqlen plugin
Corrects PillarScatter supported formats
Fix bug when reading yaml file
Signed-off-by: Shuyue Lan <shuyuel@nvidia.com>
Signed-off-by: Rajeev Rao <rajeevrao@nvidia.com>
The plugins belong in the nvinfer1::plugin namespace.
However some plugins were not only polluting the global namespace, but
were exposing the same function twice! Once in the global namespace and
once in the nvinfer1::plugin namespace. This fixes this by pushing
everything into the nvinfer1::plugin namespace.
Signed-off-by: Shuyue Lan <shuyuel@nvidia.com>
Signed-off-by: Rajeev Rao <rajeevrao@nvidia.com>
Relocate LoggerFinder to NvInferRuntime.h and make it an Iclass.
Signed-off-by: Shuyue Lan <shuyuel@nvidia.com>
Signed-off-by: Rajeev Rao <rajeevrao@nvidia.com>
Shorter plugin attribute test definitions
Fix incorrect names of some deserialize methods
Memory issues in PriorBoxPlugin
Signed-off-by: Shuyue Lan <shuyuel@nvidia.com>
Signed-off-by: Rajeev Rao <rajeevrao@nvidia.com>
- Fix vc-plugin exports to have only entry point symbols
- Add missing TENSORRTAPI for windows build
Signed-off-by: Shuyue Lan <shuyuel@nvidia.com>
Signed-off-by: Rajeev Rao <rajeevrao@nvidia.com>
Limit output_dims == W[1] for plugin test
update embLayernorm plugin doc
Remove else after return as they are not required.
Fix FlattenConcat plugin readme
Signed-off-by: Shuyue Lan <shuyuel@nvidia.com>
Signed-off-by: Rajeev Rao <rajeevrao@nvidia.com>
Kevin Chen