import itertools
import torch
import torch_npu
from torch_npu.testing.testcase import TestCase, run_tests
from torch_npu.testing.decorator import Dtypes, instantiate_tests
@instantiate_tests
class TestRandom(TestCase):
@Dtypes(torch.float32, torch.float64, torch.float16, torch.bfloat16, torch.uint8,
torch.int8, torch.int16, torch.int32, torch.int64)
def test_random_from_to(self, dtype):
size = 2000
alpha = 0.1
int64_min_val = torch.iinfo(torch.int64).min
int64_max_val = torch.iinfo(torch.int64).max
if dtype in [torch.float, torch.double, torch.half]:
min_val = int(max(torch.finfo(dtype).min, int64_min_val))
max_val = int(min(torch.finfo(dtype).max, int64_max_val))
froms = [min_val, -42, 0, 42]
tos = [-42, 42, max_val >> 1]
elif dtype == torch.bfloat16:
min_val = int64_min_val
max_val = int64_max_val
froms = [min_val, -42, 0, 42]
tos = [-42, 0, 42, max_val >> 1]
elif dtype == torch.uint8:
min_val = torch.iinfo(dtype).min
max_val = torch.iinfo(dtype).max
froms = [int64_min_val, -42, min_val - 1, min_val, 42, max_val, max_val + 1]
tos = [-42, min_val - 1, min_val, 42, max_val, max_val + 1, int64_max_val]
elif dtype == torch.int64:
min_val = int64_min_val
max_val = int64_max_val
froms = [min_val, -42, 0, 42]
tos = [-42, 42, max_val]
else:
min_val = torch.iinfo(dtype).min
max_val = torch.iinfo(dtype).max
froms = [int64_min_val, min_val - 1, min_val, -42, 0, 42, max_val, max_val + 1]
tos = [min_val - 1, min_val, 42, max_val, int64_max_val]
if dtype == torch.double:
fp_limit = 2**53
elif dtype == torch.float:
fp_limit = 2**24
elif dtype == torch.half:
fp_limit = 2**11
elif dtype == torch.bfloat16:
fp_limit = 2**8
else:
fp_limit = 0
for from_ in froms:
for to_ in tos:
self._check_random_from_to(from_, to_, dtype, min_val, max_val, fp_limit, size, alpha)
def _check_random_from_to(self, from_, to_, dtype, min_val, max_val, fp_limit, size, alpha):
t = torch.empty(size, dtype=dtype, device="cpu").to("npu")
if to_ > from_:
if not (min_val <= from_ <= max_val):
self.assertRaisesRegex(
RuntimeError,
"from is out of bounds",
lambda: t.random_(from_, to_)
)
elif not (min_val <= (to_ - 1) <= max_val):
self.assertRaisesRegex(
RuntimeError,
"to - 1 is out of bounds",
lambda: t.random_(from_, to_)
)
else:
if dtype.is_floating_point and (
not (-fp_limit <= from_ <= fp_limit) or not (-fp_limit <= (to_ - 1) <= fp_limit)):
if not (-fp_limit <= from_ <= fp_limit):
self.assertWarnsRegex(UserWarning, "from is out of bounds",
lambda: t.random_(from_, to_))
if not (-fp_limit <= (to_ - 1) <= fp_limit):
self.assertWarnsRegex(UserWarning, "to - 1 is out of bounds",
lambda: t.random_(from_, to_))
else:
t.random_(from_, to_)
range_ = to_ - from_
delta = max(1, alpha * range_)
if dtype == torch.bfloat16:
self.assertTrue(from_ <= t.to(torch.double).min() < (from_ + delta))
self.assertTrue((to_ - delta) < t.to(torch.double).max() <= to_)
else:
self.assertTrue(from_ <= t.to(torch.double).min() < (from_ + delta))
self.assertTrue((to_ - delta) <= t.to(torch.double).max() < to_)
else:
self.assertRaisesRegex(
RuntimeError,
"random_ expects 'from' to be less than 'to', but got from=" + str(from_) + " >= to=" + str(to_),
lambda: t.random_(from_, to_)
)
@Dtypes(torch.int32, torch.int64, torch.float, torch.float16)
def test_random_to(self, dtype):
size = 2000
alpha = 0.1
int64_min_val = torch.iinfo(torch.int64).min
int64_max_val = torch.iinfo(torch.int64).max
if dtype in [torch.float, torch.float16]:
min_val = int(max(torch.finfo(dtype).min, int64_min_val))
max_val = int(min(torch.finfo(dtype).max, int64_max_val))
tos = [42, max_val >> 1]
elif dtype == torch.int64:
min_val = int64_min_val
max_val = int64_max_val
tos = [42, max_val]
else:
min_val = torch.iinfo(dtype).min
max_val = torch.iinfo(dtype).max
tos = [42, max_val]
from_ = 0
for to_ in tos:
t = torch.empty(size, dtype=dtype, device="cpu").to("npu")
if to_ > from_:
if not (min_val <= (to_ - 1) <= max_val):
self.assertRaisesRegex(
RuntimeError,
"to - 1 is out of bounds",
lambda: t.random_(to_)
)
else:
t.random_(to_)
range_ = to_ - from_
delta = max(1, alpha * range_)
self.assertTrue(from_ <= t.to(torch.double).min() < (from_ + delta))
self.assertTrue((to_ - delta) <= t.to(torch.double).max() < to_)
else:
self.assertRaisesRegex(
RuntimeError,
"random_ expects 'from' to be less than 'to', but got from=" + str(from_) + " >= to=" + str(to_),
lambda: t.random_(from_, to_)
)
@Dtypes(torch.int32, torch.int64, torch.float, torch.float16, torch.uint8, torch.int16, torch.int8, torch.double)
def test_random_default(self, dtype):
size = 2000
alpha = 0.1
if dtype == torch.float:
to_inc = 1 << 24
elif dtype == torch.double:
to_inc = 1 << 53
elif dtype == torch.half:
to_inc = 1 << 11
else:
to_inc = torch.iinfo(dtype).max
t = torch.empty(size, dtype=dtype, device="cpu").to("npu")
t.random_()
self.assertTrue(0 <= t.to(torch.double).min() < alpha * to_inc)
self.assertTrue((to_inc - alpha * to_inc) < t.to(torch.double).max() <= to_inc)
@Dtypes(torch.int32, torch.int64, torch.float, torch.float16)
def test_random_diffent_size(self, dtype):
from_ = -800
to_ = 800
size = [10]
t = torch.empty(size, dtype=dtype, device="cpu").to("npu")
t.random_(from_, to_)
self.assertTrue(from_ <= t.to(torch.double).min() <= to_)
self.assertTrue(from_ <= t.to(torch.double).max() <= to_)
size = [10, 8]
t = torch.empty(size, dtype=dtype, device="cpu").to("npu")
t.random_(from_, to_)
self.assertTrue(from_ <= t.to(torch.double).min() <= to_)
self.assertTrue(from_ <= t.to(torch.double).max() <= to_)
size = [10, 8, 7]
t = torch.empty(size, dtype=dtype, device="cpu").to("npu")
t.random_(from_, to_)
self.assertTrue(from_ <= t.to(torch.double).min() <= to_)
self.assertTrue(from_ <= t.to(torch.double).max() <= to_)
size = [10, 8, 7, 5]
t = torch.empty(size, dtype=dtype, device="cpu").to("npu")
t.random_(from_, to_)
self.assertTrue(from_ <= t.to(torch.double).min() <= to_)
self.assertTrue(from_ <= t.to(torch.double).max() <= to_)
size = [10, 8, 7, 5, 2]
t = torch.empty(size, dtype=dtype, device="cpu").to("npu")
t.random_(from_, to_)
self.assertTrue(from_ <= t.to(torch.double).min() <= to_)
self.assertTrue(from_ <= t.to(torch.double).max() <= to_)
def test_random_seed(self):
torch.manual_seed(123)
input1 = torch.rand(2, 3, 4).npu()
input1.random_(2, 10)
torch.manual_seed(123)
input2 = torch.rand(2, 3, 4).npu()
input2.random_(2, 10)
self.assertRtolEqual(input1.cpu(), input2.cpu())
def test_random_seed_fp16(self):
torch.manual_seed(3)
input1 = torch.rand(4, 5, 3).half().npu()
input1.random_(0, 100)
torch.manual_seed(3)
input2 = torch.rand(4, 5, 3).half().npu()
input2.random_(0, 100)
self.assertRtolEqual(input1.cpu(), input2.cpu())
if __name__ == "__main__":
run_tests()