diff --git a/README.md b/README.md index 663bd85a..f1a3e8d3 100644 --- a/README.md +++ b/README.md @@ -40,17 +40,19 @@ python3 -m sphinx -T -E -b html -d _build/doctrees -D language=en . _build/html ### Operations - Contraction Tensor +hiptensor supports a tensor contraction of the form $D = \alpha AB + \beta C$ + Supported data-type combinations are: -| typeA | typeB | typeC | typeCompute | notes | -|-----------|-----------|-----------|-----------------|----------------------------------| -| bf16 | bf16 | bf16 | f32 | | -| __half | __half | __half | f32 | | -| f32 | f32 | f32 | bf16 | | -| f32 | f32 | f32 | __half | | -| f32 | f32 | f32 | f32 | | -| f64 | f64 | f64 | f32 | f64 is supported only on gfx90 + | -| f64 | f64 | f64 | f64 | f64 is supported only on gfx90 + | +| typeA | typeB | typeC | typeCompute | notes | +| ----------- | ----------- | ----------- | ----------------- | ---------------------------------- | +| bf16 | bf16 | bf16 | f32 | | +| __half | __half | __half | f32 | | +| f32 | f32 | f32 | bf16 | | +| f32 | f32 | f32 | __half | | +| f32 | f32 | f32 | f32 | | +| f64 | f64 | f64 | f32 | f64 is supported only on gfx90a + | +| f64 | f64 | f64 | f64 | f64 is supported only on gfx90a + | ### Operations - Permutation Tensor @@ -119,6 +121,7 @@ After configuration, build with `cmake --build -- -j` ### Logger tests Tests API implementation of logger verbosity and functionality. + ```bash /bin/logger_test ``` @@ -128,6 +131,7 @@ Tests API implementation of logger verbosity and functionality. ### Bilinear contraction tests Tests the API implementation of bilinear contraction algorithm with validation. + ```bash /bin/bilinear_contraction_test ``` @@ -135,6 +139,7 @@ Tests the API implementation of bilinear contraction algorithm with validation. ### Scale contraction tests Tests the API implementation of scale contraction algorithm with validation. + ```bash /bin/scale_contraction_test ``` @@ -143,6 +148,7 @@ Tests the API implementation of scale contraction algorithm with validation. ### Permutation tests Tests API implementation of permutation algorithm with validation. + ```bash /bin/permutation_test ``` @@ -154,6 +160,7 @@ These are stand-alone use-cases of the hipTensor contraction operations. ### F32 Bilinear contraction Demonstrates the API implementation of bilinear contraction operation without validation. + ```bash /bin/simple_bilinear_contraction_____compute_ ``` @@ -161,6 +168,7 @@ Demonstrates the API implementation of bilinear contraction operation without va ### F32 Scale contraction Demonstrates the API implementation of scale contraction operation without validation. + ```bash /bin/simple_scale_contraction____compute_ ``` @@ -168,6 +176,7 @@ Demonstrates the API implementation of scale contraction operation without valid ### Permutation Demonstrates the API implementation of permutation operation without validation. + ```bash /bin/simple_permutation ```