This repository has been archived by the owner on Jul 1, 2024. It is now read-only.
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
* New TFhub Models (#217) * adding new tfhub models to the documentation * Fix for performance degradation issue with TF2 api (#213) * Fix for performance issue with Keras api * More fixes in the parm->const logic in translate graph - enabling the env var by default - adding datatype check for tensors to create tensor data in vector - whitespace removal * Formatting fixes * Fixing TranslateGraph defn for tests * Fix for sample execution crash on GPU * GPU specific check for clean up * OVTF examples/colab samples updated to TF2.x (#214) * Update classification sample to TF2 api (saved model) * Updated object detection sample to TF2.x (saved model) * Updated README, moving TF1 classification sample to TF_1_x * Updated colab samples, requirements txt * Updated colab samples * Formatting fix * Minor fix in COlab sample * Minor documentation update * Add translations for new TF ops (#216) * Upgrade OpenVINO Opset to version 7 * Add tests for MaxPool3D, Floor * Add translations for ScatterNd, AvgPool3D * Enable Conv3DBackpropInputV2; Add Conv3D* Python tests * Update OCM * Apply Code Formatting * Disable Conv3D tests for GPU * Modify MYD tests * Update OCM * Update OCM for binary op type support issue * Update MYD Python tests list for mac * Update to OCM master * Skip redundant Conv3D tests; Disable Conv3D on MYD * Disable conv3d_transpose_test for MYD * Replace old method of pad compute in AvgPool op * Remove Conv3D Gradient Test * Readme: the introduction text is updated to include the two lines of code (#219) * Infinite loop bug fixed in dyn to static input check (#218) * Infinite loop bug fixed in dyn to static input check * Upgrade OV version to 2021.4.2 and TF version to 2.7.0 (#221) * Upgrade OV to 2021.4.2 and TF to 2.7.0 * Add fix for attributes which do not have default values in TF - 2.7 * One Gather TF test added to skip list * Remove unused yml files * Revert mac.yml to cpu-mac.yml * C++14 fix for CPP Tests * Update OV version in Mac yml file Co-authored-by: Suryaprakash Shanmugam <[email protected]> * Update model in Object Detection sample from YoloV3 to YoloV4 (#220) * Upgrade Obj Det Samples to YoloV4 * Update Code Formatting * Convert model in a virtualenv * Update Colab example with YoloV4 changes * Update examples documentation for yolov4 * Enable windows support for OVTF (#222) * Build scripts and code updated for windows support * Updated code file and ocm submoudle branch for windows support * Added default TRUE condition for wheel and framework packages for windows * OVTF Python enabled on Windows * Separate compilation of Protobuf library is disabled, required protobuf symbols are added in TF source code and the generated wheel package resolved the protobuf linking issues * C++ samples cmakelist modified to add the tensorflow C++ libs dependencies * Build script modified for windows related paths * Adding a new ngraph input/output node in encapsulate clusters is not working properly and a fix is added for the same *_ovtf_static_inputs attribute is added only if there are any static inputs for that node, as reading empty list value throws error on windows * Enabled Python and TF Python tests for Windows * Corrected a path for build on linux * Restore C++ example in CMakeList * Correction in Indentation in TF test runner to avoid build failure on Macos * Rebased with 1.0.1 release * Updated build script to copy python framework libs * Review comments incorporated, and scatter_nd python test added to skip list * Code format check applied * Change OV, TF, and OVTF versions (#223) * Change OV, TF, and OVTF versions * Remove py36 from ABI1 TF builds * New patch for TF ABI1 builds * Update manylinux2014 patch * Downgrade TF version in yolov4 conversion scripts * Refactor CI; Upgrade OCM (#228) * Update windows documentation (#229) * Update cmakelist for c++14 change infor windows * Update documentation for Windows, add test files and TF symbols patch * Update examples and its readme for windows * Apply code format * Update readme for examples and build based on review comments * Update READMEs (#230) * Update READMEs * Correction on iGPU * Add new Colab samples to the examples directory * Add Colab notebook deprecation notice * Update cmakelist for windows wheel, colab notebook links and readme files (#231) * Update Cmakelist for windows wheel, colab link in notebooks, and Readme files * Notebooks directory renamed as notebooks * Update python and tf unit tests for GPU on windows (#232) * Fix Obj Det Sample Test Cases; Fix Test Script; Minor Documentation Updates (#235) * Fix for output_layer arg * Improve documentation for readability; Fix test script issue * Code formatting * Update INSTALL.md * Backend error check added to the c++ sample (#233) * Backend error check added to the c++ sample * Modified Readme for windows release name * Updated windows release comment in main readme * Update ObjDet Conversion Scripts (#236) * Add win conversion script; Patch through git * Update documentation for Windows conv script * Updated mandarin documentation (#238) * Updated mandarin documentation * Update links in some of the mandarin readme files * Update models.md (#234) * Fix issue with tf import (#239) Co-authored-by: pratiksha123507 <[email protected]> Co-authored-by: Laxmi Ganesan <[email protected]> Co-authored-by: Suryaprakash Shanmugam <[email protected]> Co-authored-by: Mustafa Cavus <[email protected]>
- Loading branch information
5 people
authored
Dec 8, 2021
1 parent
831f1e5
commit b31f109