Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix erfinv and swapaxes #7217

Merged
merged 18 commits into from
Jan 12, 2022
Merged
Show file tree
Hide file tree
Changes from 10 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 0 additions & 3 deletions oneflow/core/functional/impl/math_functor.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -648,9 +648,6 @@ class SwapaxesFunctor {
<< "], but got " << dim_1 << ")";
return Transpose2dim(x, dim0, dim1);
}

private:
std::shared_ptr<OpExpr> op_;
};

class ArangeFunctor {
Expand Down
16 changes: 11 additions & 5 deletions oneflow/user/kernels/erfinv_kernel.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ See the License for the specific language governing permissions and
limitations under the License.
*/
#include "oneflow/core/framework/framework.h"
#include <math.h>
#define PI 3.14159265358979323846L
namespace oneflow {

template<typename T>
Expand All @@ -39,8 +39,14 @@ class CpuErfinvKernel final : public user_op::OpKernel {
T z, num, dem;
T x = x_ptr[i]; // Promise the correctness of inplace version.
T x_abs = std::abs(x);
if (x_abs > 1.0) y_ptr[i] = std::numeric_limits<T>::quiet_NaN();
if (x_abs == 1.0) y_ptr[i] = std::copysign(std::numeric_limits<T>::infinity(), x);
if (x_abs > 1.0) {
y_ptr[i] = std::numeric_limits<T>::quiet_NaN();
continue;
}
if (x_abs == 1.0) {
y_ptr[i] = std::copysign(std::numeric_limits<T>::infinity(), x);
continue;
}
if (x_abs <= static_cast<T>(central_range)) {
z = x * x;
num = (((a[3] * z + a[2]) * z + a[1]) * z + a[0]);
Expand All @@ -54,11 +60,11 @@ class CpuErfinvKernel final : public user_op::OpKernel {
}
y_ptr[i] = y_ptr[i]
- (std::erf(y_ptr[i]) - x)
/ ((static_cast<T>(2.0) / static_cast<T>(std::sqrt(M_PI)))
/ ((static_cast<T>(2.0) / static_cast<T>(std::sqrt(static_cast<double>(PI))))
* std::exp(-y_ptr[i] * y_ptr[i]));
y_ptr[i] = y_ptr[i]
- (std::erf(y_ptr[i]) - x)
/ ((static_cast<T>(2.0) / static_cast<T>(std::sqrt(M_PI)))
/ ((static_cast<T>(2.0) / static_cast<T>(std::sqrt(static_cast<double>(PI))))
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这个改动的目的是什么?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

和Pytorch的CPU版本erfinv严格对齐

* std::exp(-y_ptr[i] * y_ptr[i]));
}
}
Expand Down
26 changes: 26 additions & 0 deletions python/oneflow/test/modules/test_erfinv.py
Original file line number Diff line number Diff line change
Expand Up @@ -27,8 +27,34 @@
from oneflow.test_utils.automated_test_util import *


def _test_flow_erfinv_with_inf_data(test_case, device):
x = flow.tensor(np.ones((5, 5)), dtype=flow.float32, device=flow.device(device))
of_out = flow.erfinv(x)
np_out = np.full((5, 5), np.inf)
test_case.assertTrue(np.array_equal(of_out.numpy(), np_out))


def _test_flow_erfinv_with_nan_data(test_case, device):
x = flow.tensor(
np.arange(2, 22).reshape(4, 5), dtype=flow.float32, device=flow.device(device)
)
of_out = flow.erfinv(x)
np_out = np.full((4, 5), np.nan)
test_case.assertTrue(np.array_equal(of_out.numpy(), np_out, equal_nan=True))


@flow.unittest.skip_unless_1n1d()
class TestErfinvModule(flow.unittest.TestCase):
def test_flow_erfinv(test_case):
arg_dict = OrderedDict()
arg_dict["test_fun"] = [
_test_flow_erfinv_with_inf_data,
_test_flow_erfinv_with_nan_data,
]
arg_dict["device"] = ["cpu", "cuda"]
for arg in GenArgList(arg_dict):
arg[0](test_case, *arg[1:])

@autotest(check_graph=True, auto_backward=False)
def test_flow_erfinv_with_random_data(test_case):
device = random_device()
Expand Down