wehub-resource-sync
184 lines
6.3 KiB
Python
184 lines
6.3 KiB
Python
# Copyright (c) 2024 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 os
|
|
import unittest
|
|
|
|
import numpy
|
|
|
|
os.environ['FLAGS_prim_all'] = 'true'
|
|
os.environ['FLAGS_prim_enable_dynamic'] = 'true'
|
|
os.environ['FLAGS_use_cinn'] = '1'
|
|
os.environ['FLAGS_deny_cinn_ops'] = 'slice;'
|
|
|
|
import paddle
|
|
|
|
|
|
def generate_input_spec(rank_dtype_list):
|
|
input_spec = []
|
|
for rank, dtype in rank_dtype_list:
|
|
input_spec.append(
|
|
paddle.static.InputSpec(shape=[None] * rank, dtype=dtype)
|
|
)
|
|
return input_spec
|
|
|
|
|
|
class TestTrivialFusion(unittest.TestCase):
|
|
def setUp(self):
|
|
pass
|
|
|
|
def tearDown(self):
|
|
pass
|
|
|
|
def compare_result(self, dy_compute, input_spec, data_init):
|
|
inputs = data_init()
|
|
dy_out = dy_compute(*inputs)
|
|
static_compute = paddle.jit.to_static(
|
|
full_graph=True,
|
|
backend="CINN",
|
|
input_spec=input_spec,
|
|
)(dy_compute)
|
|
st_out = static_compute(*inputs)
|
|
if isinstance(dy_out, paddle.Tensor):
|
|
numpy.testing.assert_allclose(dy_out, st_out, atol=1e-5, rtol=1e-6)
|
|
return
|
|
for d, s in zip(dy_out, st_out):
|
|
numpy.testing.assert_allclose(d, s, atol=1e-5, rtol=1e-6)
|
|
|
|
def test_simple_trivial_fusions(self):
|
|
def func(x):
|
|
x = x * 2
|
|
x = x + 1
|
|
x = paddle.nn.functional.relu(x)
|
|
x = paddle.transpose(x, perm=[0, 2, 1])
|
|
x = x.reshape((-1, 128))
|
|
return x
|
|
|
|
def init():
|
|
x = paddle.rand((32, 32, 128))
|
|
return (x,)
|
|
|
|
input_spec = generate_input_spec([(3, 'float32')])
|
|
self.compare_result(func, input_spec, init)
|
|
|
|
def test_trivial_fusion_slice_and_concat(self):
|
|
def func(x, y):
|
|
x = x * 2
|
|
y = y * 2
|
|
x = x[:, :, :64]
|
|
y = y[:, :, :64]
|
|
z = paddle.concat([x, y], axis=-1)
|
|
return z
|
|
|
|
def init():
|
|
x = paddle.rand((32, 32, 128))
|
|
y = paddle.rand((32, 32, 128))
|
|
return (x, y)
|
|
|
|
input_spec = generate_input_spec([(3, 'float32'), (3, 'float32')])
|
|
self.compare_result(func, input_spec, init)
|
|
|
|
def test_trivial_fusion_gather_nd(self):
|
|
def func(x, y):
|
|
x = x * 2
|
|
output = paddle.gather_nd(x, y)
|
|
return output
|
|
|
|
def init():
|
|
x = paddle.to_tensor(
|
|
[[[1, 2], [3, 4], [5, 6]], [[7, 8], [9, 10], [11, 12]]]
|
|
)
|
|
index = paddle.to_tensor([[0, 1]])
|
|
return (x, index)
|
|
|
|
input_spec = [
|
|
paddle.static.InputSpec(shape=[None, None, None], dtype='float32'),
|
|
paddle.static.InputSpec(shape=[None, 2], dtype='int32'),
|
|
]
|
|
self.compare_result(func, input_spec, init)
|
|
|
|
def test_broadcast(self):
|
|
def func(x, y):
|
|
output = x + y
|
|
return output
|
|
|
|
def init():
|
|
x = paddle.rand((32, 1))
|
|
y = paddle.rand((1, 32))
|
|
return (x, y)
|
|
|
|
input_spec = generate_input_spec([(2, 'float32'), (2, 'float32')])
|
|
self.compare_result(func, input_spec, init)
|
|
|
|
def test_broadcast_tree(self):
|
|
def init():
|
|
var_1 = paddle.rand([32], dtype="float32")
|
|
var_2 = paddle.rand([32], dtype="float32")
|
|
var_3 = paddle.rand([32], dtype="float32")
|
|
return (var_1, var_2, var_3)
|
|
|
|
def input_spec():
|
|
return [
|
|
paddle.static.InputSpec(shape=[None], dtype='float32'), # S0
|
|
paddle.static.InputSpec(shape=[None], dtype='float32'), # S1
|
|
paddle.static.InputSpec(shape=[None], dtype='float32'), # S2
|
|
]
|
|
|
|
def func(var_1, var_2, var_3):
|
|
var_4 = paddle.reshape(var_1, [-1, 32]) # Div(S0, 32)
|
|
var_5 = paddle.reshape(var_2, [-1, 32]) # Div(S1, 32)
|
|
var_6 = paddle.reshape(var_3, [-1, 32]) # Div(S2, 32)
|
|
# Broadcast(Div(S0, 32), Div(S1, 32), Div(S2, 32)
|
|
var_7 = var_4 + var_5 + var_6
|
|
# Mul(Broadcast(Div(S0, 32), Div(S1, 32), Div(S2, 32)), 32)
|
|
var_9 = var_7.reshape([1, -1, 1, 1])
|
|
|
|
var_752 = paddle.full([20, var_9.shape[1], 8, 24], 0.1, "float32")
|
|
var_kwarg_var_10744 = var_2
|
|
var_769 = paddle.full(
|
|
[20, 32, var_4.shape[0], 8, 24], 0.1, "float32"
|
|
)
|
|
var_kwarg_middle_31 = paddle.rand(
|
|
[20, var_9.shape[1], 8, 24], "float32"
|
|
)
|
|
var_kwarg_middle_31[:] = 0.1
|
|
var_kwarg_middle_30 = paddle.full([20, 32, 1, 1, 1], 0.1, "float32")
|
|
|
|
var_812 = paddle.full(shape=[], dtype='float32', fill_value=0.0)
|
|
var_814 = paddle.expand(var_812, var_kwarg_middle_31.shape)
|
|
var_815 = paddle.greater_than(var_kwarg_middle_31, var_814)
|
|
var_816 = paddle.cast(var_815, dtype='float32')
|
|
var_817 = var_816 * var_752
|
|
var_818 = paddle.reshape(var_817, [20, 32, -1, 8, 24])
|
|
var_819 = paddle.reshape(var_kwarg_var_10744, [32, -1, 1, 1])
|
|
var_820 = paddle.full(shape=[20, 32, 1, 1, 1], fill_value=1e-05)
|
|
var_821 = var_kwarg_middle_30 + var_820
|
|
var_822 = paddle.full(shape=[20, 32, 1, 1, 1], fill_value=1.0)
|
|
var_823 = var_822 / var_821
|
|
var_824 = paddle.sqrt(var_823)
|
|
var_827 = var_818 * var_819
|
|
var_830 = var_824 * var_827
|
|
var_831 = paddle.sum(var_830, keepdim=True, axis=[2, 3, 4])
|
|
var_834 = var_831 * var_769
|
|
var_837 = var_834 * var_827
|
|
var_838 = paddle.sum(var_837, keepdim=True, axis=[2, 3, 4])
|
|
return var_818, var_824, var_838, var_831, var_830
|
|
|
|
self.compare_result(func, input_spec(), init)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|