wehub-resource-sync
84 lines
2.6 KiB
Python
84 lines
2.6 KiB
Python
# Copyright (c) 2025 PaddlePaddle Authors. All Rights Reserved.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
import unittest
|
|
|
|
import numpy as np
|
|
from op_test import is_custom_device
|
|
|
|
import paddle
|
|
import paddle.incubate.nn.functional as F
|
|
from paddle import core
|
|
|
|
|
|
@unittest.skipIf(
|
|
not (core.is_compiled_with_cuda() or is_custom_device()),
|
|
"core is not compiled with CUDA ",
|
|
)
|
|
class TestFusedStackTransposeQuantOp(unittest.TestCase):
|
|
def setUp(self):
|
|
np.random.seed(2025)
|
|
self.dtype = 'bfloat16'
|
|
self.transpose = True
|
|
|
|
def restore_stack_quant(self, out, scale):
|
|
# Expand scale to [M, K] shape assuming block size 128 x 128
|
|
scale = paddle.repeat_interleave(scale, repeats=128, axis=0)
|
|
scale = paddle.repeat_interleave(scale, repeats=128, axis=1)
|
|
x = out.astype('float32') * scale
|
|
return x
|
|
|
|
def check_main(self, N, M, K):
|
|
paddle.disable_static()
|
|
x_tensor_list = [
|
|
paddle.randn([M, K], dtype=self.dtype).clip(min=-50, max=50)
|
|
for _ in range(N)
|
|
]
|
|
x_fp32 = paddle.stack(x_tensor_list).reshape([-1, K]).astype('float32')
|
|
out, scale = F.fused_stack_transpose_quant(
|
|
x_tensor_list, transpose=self.transpose
|
|
)
|
|
x_restored = self.restore_stack_quant(out, scale)
|
|
if self.transpose:
|
|
x_restored = (
|
|
x_restored.reshape([N, K, M])
|
|
.transpose([0, 2, 1])
|
|
.reshape([-1, K])
|
|
)
|
|
paddle.enable_static()
|
|
|
|
if not (paddle.is_compiled_with_cuda() or is_custom_device()):
|
|
return
|
|
np.testing.assert_allclose(
|
|
x_fp32.numpy(),
|
|
x_restored.numpy(),
|
|
rtol=0.01,
|
|
atol=0.2,
|
|
)
|
|
|
|
def test_fused_stack_transpose_quant(self):
|
|
self.check_main(1, 2048, 128)
|
|
|
|
def test_fused_stack_transpose_quant2(self):
|
|
self.check_main(4, 2048, 128)
|
|
|
|
|
|
class TestFusedStackTransposeQuantOp1(TestFusedStackTransposeQuantOp):
|
|
def setUp(self):
|
|
super().setUp()
|
|
self.transpose = False
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|