Skip to content
Snippets Groups Projects
Commit 5946282b authored by zhanghb97's avatar zhanghb97
Browse files

Update README.

parent e5e0d084
No related branches found
No related tags found
No related merge requests found
......@@ -24,31 +24,32 @@ Currently, the image processing benchmark includes the following frameworks or o
Run the image processing benchmark:
*Note: Please replace the `/PATH/TO/*` with your local path.
Use `-DBUDDY_OPT_STRIP_MINING` (default: 256) and `-DBUDDY_OPT_ATTR` (default: avx512f) to specify strip-mining size and architecture extension.*
```
$ cd buddy-benchmark
$ mkdir build && cd build
$ cmake -G Ninja .. \
-DIMAGE_PROCESSING_BENCHMARKS=ON \
-DOpenCV_DIR=/path/to/opencv/build/ \
-DBUDDY_OPT_BUILD_DIR=/path/to/buddy-mlir/build/ \
-DBUDDY_OPT_STRIP_MINING=<strip mining size, default: 256> \
-DBUDDY_OPT_ATTR=<ISA vector extension, default: avx512f>
-DOpenCV_DIR=/PATH/TO/OPENCV/BUILD/ \
-DBUDDY_OPT_BUILD_DIR=/PATH/TO/BUDDY-MLIR/BUILD/
$ ninja image-processing-benchmark
$ cd bin && ./image-processing-benchmark
```
Note : The convolution implementation in buddy mlir is not feature complete at the moment and it may produce output which differs to some extent from the frameworks used in comparison.
## Deep Learning Benchmark
*Note: Please replace the `/PATH/TO/*` with your local path.
Use `-DBUDDY_OPT_ATTR` (default: avx512f) to specify architecture extension.*
```
$ cd buddy-benchmark
$ mkdir build && cd build
$ cmake -G Ninja .. \
-DDEEP_LEARNING_BENCHMARKS=ON \
-DOpenCV_DIR=/path/to/opencv/build/ \
-DBUDDY_OPT_BUILD_DIR=/path/to/buddy-mlir/build/ \
-DBUDDY_OPT_ATTR=<ISA vector extension, default: avx512f>
-DOpenCV_DIR=/PATH/TO/OPENCV/BUILD/ \
-DBUDDY_OPT_BUILD_DIR=/PATH/TO/BUDDY-MLIR/BUILD/
$ ninja
```
......@@ -56,7 +57,7 @@ The deep learning benchmark includes the following e2e models and operations:
- MobileNet
NOTE: We generated the model code with IREE and made appropriate modifications, and then compiled it with the MLIR tool chain.
We generated the model code with IREE and made appropriate modifications, and then compiled it with the MLIR tool chain.
Run the MobileNet benchmark:
......
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment