Skip to content
This repository has been archived by the owner on Jan 24, 2024. It is now read-only.

Commit

Permalink
op unittest for clz/popc (#1498)
Browse files Browse the repository at this point in the history 
* op unittest for popc

* refine clz test
  • Loading branch information
@zzk0
zzk0 authored Jun 5, 2023
1 parent 1a928df commit 30b9779
Show file tree
Hide file tree
Showing 2 changed files with 183 additions and 114 deletions.
148 changes: 92 additions & 56 deletions python/tests/ops/test_clz_op.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -14,35 +14,55 @@
# See the License for the specific language governing permissions and
# limitations under the License.

import unittest
import numpy as np
from op_test import OpTest, OpTestTool
import paddle
import cinn
from cinn.frontend import *
from cinn.common import *
from cinn.frontend import *
from op_test import OpTest, OpTestTool
from op_test_helper import TestCaseHelper

INT32_MAX = (1 << 31) - 1
INT32_MIN = -(1 << 31)
INT64_MAX = (1 << 63) - 1
INT64_MIN = -(1 << 63)


def count_leading_zeros(integer, dtype):
if dtype == "int32":
bits = 32
elif dtype == "int64":
bits = 64
else:
raise NotImplementedError
if integer < 0:
return 0
mask = 1 << (bits - 1)
integer &= (mask - 1)
clz = 0
while mask > 0 and integer & mask == 0:
clz += 1
mask >>= 1
return clz


@OpTestTool.skip_if(not is_compiled_with_cuda(),
"x86 test will be skipped due to timeout.")
class TestClzOp(OpTest):
def setUp(self):
self.init_case()
print(f"\nRunning {self.__class__.__name__}: {self.case}")
self.inputs = {}
self.prepare_inputs()

def init_case(self):
self.inputs = {
# "x": self.random([32, 64], 'int32', low = -2147483648, high=2147483647)
"x":
np.array([
-1591895863, -1770335025, -1290313501, 478042597, 189030958,
-935228100, 718518127, -2066013593, -1028229638, -1930307001,
-858478166, -282304333
]).astype(np.int32)
}
self.outputs = {
"y": np.array([0, 0, 0, 3, 4, 0, 2, 0, 0, 0, 0,
0]).astype(np.int32)
}
def prepare_inputs(self):
dtype = self.case["dtype"]
low = INT32_MIN if dtype == "int32" else INT64_MIN
high = INT32_MAX if dtype == "int32" else INT64_MAX
x = self.random(self.case["shape"], dtype, low=low, high=high)
y = list(
map(lambda num: count_leading_zeros(num, dtype),
x.reshape(-1).tolist()))
self.inputs = {"x": x}
self.outputs = {"y": np.array(y).reshape(x.shape).astype(dtype)}

def build_paddle_program(self, target):
y = paddle.to_tensor(self.outputs["y"], stop_gradient=False)
Expand All @@ -57,48 +77,64 @@ def build_cinn_program(self, target):
prog = builder.build()
res = self.get_cinn_output(prog, target, [x], [self.inputs["x"]],
[out])
self.cinn_outputs = [res[0]]
self.cinn_outputs = res

def test_check_results(self):
self.check_outputs_and_grads()


class TestClzCase1(TestClzOp):
def init_case(self):
self.inputs = {
# "x": self.random([48, 36], 'int32', low = -2147483648, high=2147483647)
"x":
np.array([[
-780762106, 2088944770, 1793870564, 995233974, -1566864405,
-1550063384
],
[
58189437, -585656506, 1058816786, -1676158651,
-175192886, 2129254990
]]).astype(np.int32)
}
self.outputs = {
"y":
np.array([[0, 1, 1, 2, 0, 0], [6, 0, 2, 0, 0, 1]]).astype(np.int32)
}


class TestClzCase2(TestClzOp):
def init_case(self):
self.inputs = {
# "x": self.random([4, 3, 5, 8], 'int64', low = -9223372036854775808, high=9223372036854775807)
"x":
np.array([
-2603587548323400654, 5370659515557365091,
-2051413160116828951, 9015154622229049624,
-8328245342679021727, -8113334794330105534,
7187230222985732039, 1835610600500058242
]).astype(np.int64)
}
self.outputs = {
"y": np.array([0, 1, 0, 1, 0, 0, 1, 3]).astype(np.int64)
}
class TestClzOpShapeDtype(TestCaseHelper):
def init_attrs(self):
self.class_name = "TestClzOpShapeDtype"
self.cls = TestClzOp
self.inputs = [
{
"shape": [10],
},
{
"shape": [8, 5],
},
{
"shape": [10, 3, 5],
},
{
"shape": [80, 40, 5, 7],
},
{
"shape": [80, 1, 5, 7],
},
{
"shape": [80, 3, 1024, 7],
},
{
"shape": [10, 5, 2048, 2],
},
{
"shape": [1],
},
{
"shape": [512],
},
{
"shape": [1024],
},
{
"shape": [2048],
},
{
"shape": [1, 1, 1, 1],
},
]
self.dtypes = [
{
"dtype": "int32"
},
{
"dtype": "int64"
},
]
self.attrs = []


if __name__ == "__main__":
unittest.main()
TestClzOpShapeDtype().run()
149 changes: 91 additions & 58 deletions python/tests/ops/test_popc_op.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -14,36 +14,54 @@
# See the License for the specific language governing permissions and
# limitations under the License.

import unittest
import numpy as np
from op_test import OpTest, OpTestTool
import paddle
import cinn
from cinn.frontend import *
from cinn.common import *
from cinn.frontend import *
from op_test import OpTest, OpTestTool
from op_test_helper import TestCaseHelper

INT32_MAX = (1 << 31) - 1
INT32_MIN = -(1 << 31)
INT64_MAX = (1 << 63) - 1
INT64_MIN = -(1 << 63)


def popcount(integer, dtype):
if dtype == "int32":
bits = 32
elif dtype == "int64":
bits = 64
else:
raise NotImplementedError
ones = bin(integer).count("1")
if integer > 0:
return ones
else:
trailing_zeros = 0
mask = 1
while trailing_zeros < bits and ((integer & mask) == 0):
trailing_zeros += 1
mask <<= 1
return bits - ones - trailing_zeros + 1


@OpTestTool.skip_if(not is_compiled_with_cuda(),
"x86 test will be skipped due to timeout.")
class TestPopcOp(OpTest):
def setUp(self):
self.init_case()
print(f"\nRunning {self.__class__.__name__}: {self.case}")
self.inputs = {}
self.prepare_inputs()

def init_case(self):
self.inputs = {
# "x": self.random([32, 64], 'int32', low = -2147483648, high=2147483647)
"x":
np.array([
-1591895863, -1770335025, -1290313501, 478042597, 189030958,
-935228100, 718518127, -2066013593, -1028229638, -1930307001,
-858478166, -282304333
]).astype(np.int32)
}
self.outputs = {
"y":
np.array([14, 19, 16, 18, 12, 13, 20, 15, 19, 17, 16,
17]).astype(np.int32)
}
def prepare_inputs(self):
dtype = self.case["dtype"]
low = INT32_MIN if dtype == "int32" else INT64_MIN
high = INT32_MAX if dtype == "int32" else INT64_MAX
x = self.random(self.case["shape"], dtype, low=low, high=high)
y = list(map(lambda num: popcount(num, dtype), x.reshape(-1).tolist()))
self.inputs = {"x": x}
self.outputs = {"y": np.array(y).reshape(x.shape).astype(dtype)}

def build_paddle_program(self, target):
y = paddle.to_tensor(self.outputs["y"], stop_gradient=False)
Expand All @@ -58,49 +76,64 @@ def build_cinn_program(self, target):
prog = builder.build()
res = self.get_cinn_output(prog, target, [x], [self.inputs["x"]],
[out])
self.cinn_outputs = [res[0]]
self.cinn_outputs = res

def test_check_results(self):
self.check_outputs_and_grads()


class TestPopcCase1(TestPopcOp):
def init_case(self):
self.inputs = {
# "x": self.random([48, 36], 'int32', low = -2147483648, high=2147483647)
"x":
np.array([[
-780762106, 2088944770, 1793870564, 995233974, -1566864405,
-1550063384
],
[
58189437, -585656506, 1058816786, -1676158651,
-175192886, 2129254990
]]).astype(np.int32)
}
self.outputs = {
"y":
np.array([[13, 12, 16, 14, 18, 17], [19, 18, 14, 16, 17,
20]]).astype(np.int32)
}


class TestPopcCase2(TestPopcOp):
def init_case(self):
self.inputs = {
# "x": self.random([4, 3, 5, 8], 'int64', low = -9223372036854775808, high=9223372036854775807)
"x":
np.array([
-2603587548323400654, 5370659515557365091,
-2051413160116828951, 9015154622229049624,
-8328245342679021727, -8113334794330105534,
7187230222985732039, 1835610600500058242
]).astype(np.int64)
}
self.outputs = {
"y": np.array([34, 32, 34, 32, 29, 31, 34, 32]).astype(np.int64)
}
class TestPopcOpShapeDtype(TestCaseHelper):
def init_attrs(self):
self.class_name = "TestPopcOpShapeDtype"
self.cls = TestPopcOp
self.inputs = [
{
"shape": [10],
},
{
"shape": [8, 5],
},
{
"shape": [10, 3, 5],
},
{
"shape": [80, 40, 5, 7],
},
{
"shape": [80, 1, 5, 7],
},
{
"shape": [80, 3, 1024, 7],
},
{
"shape": [10, 5, 2048, 2],
},
{
"shape": [1],
},
{
"shape": [512],
},
{
"shape": [1024],
},
{
"shape": [2048],
},
{
"shape": [1, 1, 1, 1],
},
]
self.dtypes = [
{
"dtype": "int32"
},
{
"dtype": "int64"
},
]
self.attrs = []


if __name__ == "__main__":
unittest.main()
TestPopcOpShapeDtype().run()

0 comments on commit 30b9779

Please sign in to comment.