Merge pull request #239 from KeDengMS/kedeng/gelu
Fixes to Gelu for half and fusion
This commit is contained in:
commit
f58b843951
@ -139,7 +139,7 @@ struct GELU {
|
||||
CUTLASS_HOST_DEVICE
|
||||
T operator()(T const &scalar) const {
|
||||
return T(cutlass::constants::half<T>() * scalar *
|
||||
(cutlass::constants::one<T>() + erff( scalar / cutlass::constants::root_two<T>() )));
|
||||
(cutlass::constants::one<T>() + (T)erff((float)(scalar / cutlass::constants::root_two<T>()))));
|
||||
}
|
||||
};
|
||||
|
||||
@ -152,6 +152,15 @@ struct GELU<float> {
|
||||
}
|
||||
};
|
||||
|
||||
template <>
|
||||
struct GELU<double> {
|
||||
CUTLASS_HOST_DEVICE
|
||||
double operator()(double const &scalar) const {
|
||||
return cutlass::constants::half<double>() * scalar *
|
||||
(cutlass::constants::one<double>() + erf( scalar / cutlass::constants::root_two<double>() ));
|
||||
}
|
||||
};
|
||||
|
||||
template <typename T, int N>
|
||||
struct GELU<Array<T, N> > {
|
||||
CUTLASS_HOST_DEVICE
|
||||
|
@ -133,7 +133,8 @@ public:
|
||||
|
||||
/// Functionally required for serial reduction in the epilogue
|
||||
CUTLASS_HOST_DEVICE
|
||||
void set_k_partition(int k_partition) {
|
||||
void set_k_partition(int k_partition, int k_partition_count) {
|
||||
CUTLASS_UNUSED(k_partition_count);
|
||||
if (k_partition) {
|
||||
beta_ = ElementCompute(1);
|
||||
}
|
||||
|
@ -29,6 +29,8 @@
|
||||
#include "../../common/cutlass_unit_test.h"
|
||||
|
||||
#include "cutlass/epilogue/thread/linear_combination.h"
|
||||
#include "cutlass/epilogue/thread/linear_combination_gelu.h"
|
||||
#include "cutlass/epilogue/thread/activation.h"
|
||||
|
||||
/////////////////////////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
@ -119,3 +121,40 @@ TEST(Epilogue_thread_linear_combination, device_side_f16_f32_ptr) {
|
||||
}
|
||||
|
||||
/////////////////////////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
TEST(Epilogue_thread_linear_combination_gelu, device_side_f16_f16_ptr) {
|
||||
|
||||
using Element = cutlass::half_t;
|
||||
using ElementOutput = cutlass::half_t;
|
||||
int const kCount = 8;
|
||||
|
||||
using LinearCombination = cutlass::epilogue::thread::LinearCombinationGELU<
|
||||
ElementOutput,
|
||||
kCount,
|
||||
Element,
|
||||
Element>;
|
||||
|
||||
Element alpha = Element(1);
|
||||
Element beta = Element(0);
|
||||
|
||||
typename LinearCombination::Params params(&alpha, &beta);
|
||||
|
||||
LinearCombination linear_combination_op(params);
|
||||
|
||||
cutlass::Array<Element, kCount> accum;
|
||||
|
||||
for (int i = 0; i < kCount; ++i) {
|
||||
accum[i] = Element((float)i * 0.3f);
|
||||
}
|
||||
|
||||
cutlass::Array<ElementOutput, kCount> destination = linear_combination_op(accum, accum);
|
||||
cutlass::epilogue::thread::GELU<ElementOutput> gelu_func;
|
||||
|
||||
for (int i = 0; i < kCount; ++i) {
|
||||
ElementOutput expected = gelu_func(accum[i]);
|
||||
ElementOutput got = destination[i];
|
||||
EXPECT_TRUE(expected == got);
|
||||
}
|
||||
}
|
||||
|
||||
/////////////////////////////////////////////////////////////////////////////////////////////////
|
Loading…
Reference in New Issue
Block a user