* Support cblas library in dense
* start to add support for generic batch_matmul compute
* Add x86 override for batch_matmul
* Fix linting
* reset file
* Fix typos
* dummy change to re-trigger CI
* [Relay][VM]VM debugger
* Report mean/min/max for op duration
* Typos
* Lint
* Lint
* Lint
* Support build debug VM in CMake
* Lint
* Enable VM debug in unit test
* Disable debug vm test until new docker image is built
* Add device sync code
* Fix qnn unit test
* Disable vm debug by default
* Rename files
* Rename classes
* Fix comment
* Fix comment
* transpose implementation for tflite.py
* add TRANSPOSE to convert_map
* Fix Unexpected keyword argument 'axis' in function call
* add test for transpose oprator
* Add the parameter 'axes' handling
* add test for transpose oprator
* solve conflict within CONTRIBUTORS.md
* Improve the if condition for empty tuple
* Add one unit test to cover empty tuple
* solve conflict within CONTRIBUTORS.md
* QNN quantize and dequantize operators.
* addressing review comments.
* addressing review comments.
* Adding new line at the end of the file.
* Adhering to styling guidelines.
* Adding name to contributors.
* Fixing lint issue.
* Fixing file name.
* Removing unnecessary code.
* Support BatchMatMul with shapes greater than length 3
* Fixes
* Add tests
* Remove dependency on Python3
* Clean up
* Merge with master
* Resolve comments
* [Relay] [Quantization] WIP - Common files for the qauntization work.
* [Relay] [Quantization] WIP - Prototyping requantize op.
* Requantize operator implementation.
Requantize converts one quantized tensor representation to another quantized
representation. The PR has following implementation features
- Requantize operator defined in qnn namespace - relay.qnn.requantize
- Lowering of the requantize to exisiting Relay operators
- Integer fixed point implementation of requantize
- Two rounding modes - FE_UPWARDS (round towards infinity) and
FE_AWAY_FROM_ZERO (std::round behavior)
- Floating point implementation as well, that can act as reference or can be
used for devices when FP32 computation is not used.
- Unit test cases
Relevant Issue - https://github.com/dmlc/tvm/issues/2351
Credit to TFLite and GemmLowp to provide reference implementations.
* Typo and lint fixes.
* Doc fix.
* Uncommenting the lint script (fixing mistake).
* Modifying the unit tests.
* Moving C++ files into src/relay/qnn
* Moving python files to python/tvm/relay/qnn. Some minor fixes.
* Moving the attrs.h inside the include directory.
* Pushing files that I forgot earlier. Changing util location.
* Incorporating comments. API change. Lint fixes.
* Modifying the GetFixedPointMultiplierShift API as per comments.
* Forgot the dialect change.
* Changing rewrite to qnn_lower.
* Renaming Quantize to Qnn for clarity.
* Remove use_int_domain.
* Incorportaing review comments.
* Adding API doc for QNN dialect.
* Move the qnn_lower pass to transform namespace.
* Moving from expr to module. Adding namespace in C++.
* Minor sentence rewrites. Added qnn namespace.
* Added the API doc.
* Chanding default out_dtype to int8. Adding a test with in/out_dtype as uint8.
* Style fixes. Better error messages.
* Adding documentation.
* More documentation fixes.
* Adding out dtype check for requantize.
* Adding corner case for FP32 to fixed point conversion.
* Adding extra line.
* Documentation fix.
* Adding static inline.
* Incorporating jackwish comment. Removed idtype from requantize lowering.
* Removing Quantize/Dequantize code. Restricting Requantize to (u)int8/int32.
* Style fixes.
* Fix the docs.
* Move to Legalize API.
* [Relay] Rewrite pass.
This pass transforms an expression to other expression.
This pass has many usecases
* Replace a expr to another expr, if the other expr has faster performance.
* For ASICs, we might want to modify the inputs to adapt to the HW support.
* Alter op layout can work in conjunction with this pass.
The supporting usecase is the Intel i8 x i8 conv. Intel HW supports u8 x i8 conv
in HW. Using this pass, we can replace an i8 x i8 conv to a sequence of
operators where one of the operators is now u8 x i8 conv. This will also help
automatic quantizaion performance.
* Better API name.
* Removing the conv2d legalization for x86. Will send a separate PR.
* Test name changes.
* Registering one funtion to register FTVMLegalize.
* Better comments.
Neo Chien
Haichen Shen
Josh Fromm
SWu
Alexander Pivovarov
Animesh Jain
Wuwei Lin
lixiaoquan
Yong Wu
雾雨魔理沙
Jon Soifer
Wei Chen
Zhao Wu
ziheng
shoubhik
Zhi
Yulun Yao
Wuwei Lin
alexgl-github
Yifan Xiong