Skip to content

【Hackthon 6th No. 10】Upgrade isinf to support int8 int16 uint8 -part #63222

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.

Sign up for GitHub

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

Merged
luotao1 merged 2 commits into from
Apr 7, 2024
Merged

【Hackthon 6th No. 10】Upgrade isinf to support int8 int16 uint8 -part #63222

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a254793
upgrade isinf to support int8 int16 uint8
NKNaN Apr 3, 2024
7e64769
update dtype
NKNaN Apr 5, 2024
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
5 changes: 4 additions & 1 deletion paddle/phi/kernels/cpu/isfinite_kernel.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,10 @@ PD_REGISTER_KERNEL(isinf,
phi::dtype::float16,
phi::dtype::bfloat16,
int,
int64_t) {
int64_t,
int16_t,
int8_t,
uint8_t) {
kernel->OutputAt(0).SetDataType(phi::DataType::BOOL);
}

Expand Down
5 changes: 4 additions & 1 deletion paddle/phi/kernels/gpu/isfinite_kernel.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,10 @@ PD_REGISTER_KERNEL(isinf,
phi::dtype::float16,
phi::dtype::bfloat16,
int,
int64_t) {
int64_t,
int16_t,
int8_t,
uint8_t) {
kernel->OutputAt(0).SetDataType(phi::DataType::BOOL);
}

Expand Down
5 changes: 4 additions & 1 deletion python/paddle/tensor/math.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4400,7 +4400,7 @@ def isinf(x, name=None):
Return whether every element of input tensor is `+/-INF` or not.

Args:
x (Tensor): The input tensor, it's data type should be float16, float32, float64, int32, int64.
x (Tensor): The input tensor, it's data type should be float16, float32, float64, uint8, int8, int16, int32, int64.
name (str, optional): Name for the operation (optional, default is None). For more information, please refer to :ref:`api_guide_Name`.

Returns:
Expand Down Expand Up @@ -4428,8 +4428,11 @@ def isinf(x, name=None):
'float16',
'float32',
'float64',
'int8',
'int16',
'int32',
'int64',
'uint8',
'uint16',
],
'isinf',
Expand Down
34 changes: 32 additions & 2 deletions test/legacy_test/test_isfinite_v2_op.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -111,9 +111,33 @@ def np_data_generator(
},
]

TEST_META_DATA_ADDITIONAL = [
{
'low': 0.1,
'high': 1,
'np_shape': [2, 3, 4, 5],
'type': 'int8',
'sv_list': [np.inf, np.nan],
},
{
'low': 0,
'high': 100,
'np_shape': [11, 17, 10],
'type': 'int16',
'sv_list': [np.inf, np.nan],
},
{
'low': 0,
'high': 999,
'np_shape': [132],
'type': 'uint8',
'sv_list': [np.inf, np.nan],
},
]


def test(test_case, op_str, use_gpu=False):
for meta_data in TEST_META_DATA:
def test(test_case, op_str, use_gpu=False, data_set=TEST_META_DATA):
for meta_data in data_set:
meta_data = dict(meta_data)
meta_data['op_str'] = op_str
x_np, result_np = np_data_generator(**meta_data)
Expand Down Expand Up @@ -144,6 +168,9 @@ def test_nan(self):
def test_finite(self):
test(self, 'isfinite')

def test_inf_additional(self):
test(self, 'isinf', data_set=TEST_META_DATA_ADDITIONAL)


class TestCUDANormal(unittest.TestCase):
def test_inf(self):
Expand All @@ -155,6 +182,9 @@ def test_nan(self):
def test_finite(self):
test(self, 'isfinite', True)

def test_inf_additional(self):
test(self, 'isinf', True, data_set=TEST_META_DATA_ADDITIONAL)


class TestError(unittest.TestCase):
@test_with_pir_api
Expand Down