1. b6eb353 COMPMID-1478: Stop relying on static default OpenCL objects in cl2.hpp by Anthony Barbier · 6 years ago
  2. 6200fa4 COMPMID-1288 Optimizing CLGEMMLowp using 8 bit dot product instruction by Giorgio Arena · 6 years ago
  3. 76c8564 COMPMID-1083 : Compute library should be made usable on non-ARM platforms by Vidhya Sudhan Loganathan · 6 years ago
  4. f1f4906 COMPMID-655 : Check FP16 is supported by the GPU by Vidhya Sudhan Loganathan · 6 years ago
  5. ef9e059 COMPMID-1122: Don't try to retrieve the list of devices from a null context by Anthony Barbier · 6 years ago
  6. 4dcb583 COMPMID-1122: Store the cl_context only in one place by Anthony Barbier · 6 years ago
  7. 48c19f1 COMPMID-959 Refactor OpenCL interceptors to use lambda functions by Anthony Barbier · 6 years ago
  8. 7da55aa COMPMID-959: Add accessors for the OpenCL program cache by Anthony Barbier · 6 years ago
  9. c357c47 COMPMID-1008: Fix Doxygen issues by Alex Gilday · 6 years ago
  10. 847864d COMPMID-995 Add CL_DEVICE_VERSION to the test framework output by Anthony Barbier · 6 years ago
  11. 200b6e3 COMPMID-839 Added method to clear the Kernel Library's program cache by Anthony Barbier · 6 years ago
  12. f202e50 COMPMID-556 Improved indentation and error handling in format_doxygen.py by Anthony Barbier · 7 years ago
  13. f450caa COMPMID-661: softmax-uint8 implementation (#16) by Chunosov · 7 years ago
  14. d621bca COMPMID-661: directconv-uint8 (#20) by Chunosov · 7 years ago
  15. 388d3ec COMPMID-556: Support beta for all softmax data types. by Georgios Pinitas · 7 years ago
  16. 5a6e053 COMPUTE-8024 Fixed the maximum OpenCL workgroup size by Abel Bernabeu · 7 years ago
  17. 5f91072 COMPMID-513 Choose maximum local workgroup size at run time by steniu01 · 7 years ago
  18. d7e8281 COMPMID-408 Create OpenCL complex math functions for 8 bit fixed point arithmetic. by Michalis Spyrou · 7 years ago
  19. 6ff3b19 COMPMID-344 Updated doxygen by Anthony Barbier · 7 years ago