wehub-resource-sync
509 lines
17 KiB
Python
509 lines
17 KiB
Python
# Copyright (c) 2021 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 OpTest, get_device_place
|
|
|
|
import paddle
|
|
import paddle.incubate.nn.functional as incubate_f
|
|
import paddle.nn.functional as F
|
|
from paddle.nn.layer import transformer
|
|
from paddle.nn.layer.common import Dropout, Linear
|
|
from paddle.nn.layer.norm import LayerNorm
|
|
|
|
|
|
class TestFusedFFNOp(OpTest):
|
|
def getDtype(self):
|
|
self.dtype = "float32"
|
|
self.layer_norm_dtype = "float32"
|
|
|
|
def getShape(self):
|
|
self.batch_size = np.random.randint(1, 32)
|
|
self.query_length = np.random.randint(32, 128)
|
|
self.d_model = np.random.randint(32, 512)
|
|
self.dim_feedforward = np.random.randint(32, 512)
|
|
|
|
def getDiff(self):
|
|
self.rtol = 1e-3
|
|
# FIXME(limin29): Because there is a problem with the test precision
|
|
# on A100, atol is temporarily set to 1e-2, and it will be
|
|
# changed back after the precision problem is solved.
|
|
self.atol = 1e-2
|
|
if "V100" in paddle.device.cuda.get_device_name():
|
|
self.atol = 1e-4
|
|
|
|
def getActivation(self):
|
|
self.act_method = "gelu"
|
|
|
|
def getNormalizeBefore(self):
|
|
self.pre_layer_norm = False
|
|
|
|
def setUp(self):
|
|
paddle.disable_static()
|
|
self.__class__.op_type = "fused_feedforward"
|
|
# check grad in test_out_and_grad()
|
|
self.__class__.no_need_check_grad = True
|
|
self.getDtype()
|
|
self.getShape()
|
|
self.getDiff()
|
|
self.getActivation()
|
|
self.getNormalizeBefore()
|
|
paddle.set_default_dtype(self.dtype)
|
|
self.weight_attr = None
|
|
self.bias_attr = None
|
|
|
|
self.weight_attrs = transformer._convert_param_attr_to_list(
|
|
self.weight_attr, 2
|
|
)
|
|
self.bias_attrs = transformer._convert_param_attr_to_list(
|
|
self.bias_attr, 2
|
|
)
|
|
self.linear1 = Linear(
|
|
self.d_model,
|
|
self.dim_feedforward,
|
|
self.weight_attrs[1],
|
|
bias_attr=self.bias_attrs[1],
|
|
)
|
|
self.linear2 = Linear(
|
|
self.dim_feedforward,
|
|
self.d_model,
|
|
self.weight_attrs[1],
|
|
bias_attr=self.bias_attrs[1],
|
|
)
|
|
|
|
paddle.set_default_dtype(self.layer_norm_dtype)
|
|
self.norm1 = LayerNorm(self.d_model)
|
|
self.norm2 = LayerNorm(self.d_model)
|
|
self.dropout = Dropout(0.0, mode="upscale_in_train")
|
|
self.dropout1 = Dropout(0.0, mode="upscale_in_train")
|
|
self.dropout2 = Dropout(0.0, mode="upscale_in_train")
|
|
self.activation = getattr(F, self.act_method)
|
|
|
|
self.src = np.random.random(
|
|
(self.batch_size, self.query_length, self.d_model)
|
|
).astype(self.dtype)
|
|
self.dout = np.random.random(
|
|
(self.batch_size, self.query_length, self.d_model)
|
|
).astype(self.dtype)
|
|
|
|
def Base(self):
|
|
paddle.disable_static()
|
|
tensor_src = paddle.to_tensor(self.src, stop_gradient=False)
|
|
residual = tensor_src
|
|
if self.pre_layer_norm:
|
|
ln1_out = self.norm1(tensor_src)
|
|
linear2_out = self.linear2(
|
|
self.dropout(self.activation(self.linear1(ln1_out)))
|
|
)
|
|
dropout2_out = residual + self.dropout2(linear2_out)
|
|
paddle.autograd.backward(
|
|
[dropout2_out], [paddle.to_tensor(self.dout)], True
|
|
)
|
|
return dropout2_out, tensor_src.grad
|
|
else:
|
|
linear2_out = self.linear2(
|
|
self.dropout(self.activation(self.linear1(tensor_src)))
|
|
)
|
|
dropout2_out = residual + self.dropout2(linear2_out)
|
|
dropout2_out = self.norm2(dropout2_out)
|
|
paddle.autograd.backward(
|
|
[dropout2_out], [paddle.to_tensor(self.dout)], True
|
|
)
|
|
return dropout2_out, tensor_src.grad
|
|
|
|
def FusedFFN(self):
|
|
paddle.disable_static()
|
|
linear1_weight = paddle.to_tensor(
|
|
self.linear1.weight, stop_gradient=False
|
|
)
|
|
linear1_bias = paddle.to_tensor(self.linear1.bias, stop_gradient=False)
|
|
linear2_weight = paddle.to_tensor(
|
|
self.linear2.weight, stop_gradient=False
|
|
)
|
|
linear2_bias = paddle.to_tensor(self.linear2.bias, stop_gradient=False)
|
|
ln1_scale = paddle.to_tensor(self.norm1.weight, stop_gradient=False)
|
|
ln1_bias = paddle.to_tensor(self.norm1.bias, stop_gradient=False)
|
|
ln2_scale = paddle.to_tensor(self.norm2.weight, stop_gradient=False)
|
|
ln2_bias = paddle.to_tensor(self.norm2.bias, stop_gradient=False)
|
|
x = paddle.to_tensor(self.src, stop_gradient=False)
|
|
out = incubate_f.fused_feedforward(
|
|
x,
|
|
linear1_weight,
|
|
linear2_weight,
|
|
linear1_bias,
|
|
linear2_bias,
|
|
ln1_scale,
|
|
ln1_bias,
|
|
ln2_scale,
|
|
ln2_bias,
|
|
0.0,
|
|
0.0,
|
|
activation=self.act_method,
|
|
pre_layer_norm=self.pre_layer_norm,
|
|
)
|
|
paddle.autograd.backward([out], [paddle.to_tensor(self.dout)])
|
|
return out, x.grad
|
|
|
|
def test_out_and_grad(self):
|
|
paddle.seed(42)
|
|
base_out, base_grad = self.Base()
|
|
fused_out, fused_grad = self.FusedFFN()
|
|
np.testing.assert_allclose(
|
|
base_out.numpy(), fused_out.numpy(), rtol=self.rtol, atol=self.atol
|
|
)
|
|
np.testing.assert_allclose(
|
|
base_grad.numpy(),
|
|
fused_grad.numpy(),
|
|
rtol=self.rtol,
|
|
atol=self.atol,
|
|
)
|
|
|
|
|
|
class TestFusedFFNOpFp16(TestFusedFFNOp):
|
|
def getDtype(self):
|
|
self.dtype = "float16"
|
|
self.layer_norm_dtype = "float32"
|
|
|
|
def getDiff(self):
|
|
self.rtol = 1e-1
|
|
self.atol = 1e-2
|
|
|
|
def getShape(self):
|
|
self.batch_size = 4
|
|
self.query_length = 32
|
|
self.d_model = 128
|
|
self.dim_feedforward = 256
|
|
|
|
|
|
class TestFusedFFNOpFp64(TestFusedFFNOp):
|
|
def getDtype(self):
|
|
self.dtype = "float64"
|
|
self.layer_norm_dtype = "float64"
|
|
|
|
|
|
class TestFusedFFNOpActivation(TestFusedFFNOp):
|
|
def getActivation(self):
|
|
self.act_method = "relu"
|
|
|
|
|
|
class TestFusedFFNOpNormalizeBefore(TestFusedFFNOp):
|
|
def getNormalizeBefore(self):
|
|
self.pre_layer_norm = True
|
|
|
|
def getShape(self):
|
|
self.batch_size = 1
|
|
self.query_length = 1
|
|
self.d_model = 8
|
|
self.dim_feedforward = 8
|
|
|
|
|
|
class APITestStaticFusedFFN(unittest.TestCase):
|
|
def setUp(self):
|
|
self.dtype = "float32"
|
|
self.layer_norm_dtype = "float32"
|
|
self.batch_size = 1
|
|
self.d_model = 8
|
|
self.dim_feedforward = 8
|
|
|
|
def run_fused_feedforward(
|
|
self,
|
|
x_data,
|
|
linear1_weight_data,
|
|
linear1_bias_data,
|
|
linear2_weight_data,
|
|
linear2_bias_data,
|
|
ln1_scale_data,
|
|
ln1_bias_data,
|
|
ln2_scale_data,
|
|
ln2_bias_data,
|
|
):
|
|
main = paddle.static.Program()
|
|
startup = paddle.static.Program()
|
|
paddle.seed(42)
|
|
with paddle.static.program_guard(main, startup):
|
|
x = paddle.static.data(
|
|
name='x',
|
|
shape=[self.batch_size, self.d_model, self.dim_feedforward],
|
|
dtype=self.dtype,
|
|
)
|
|
linear1_weight = paddle.static.data(
|
|
name='linear1_weight',
|
|
shape=[self.d_model, self.dim_feedforward],
|
|
dtype=self.dtype,
|
|
)
|
|
linear1_bias = paddle.static.data(
|
|
name='linear1_bias', shape=[self.dim_feedforward]
|
|
)
|
|
linear2_weight = paddle.static.data(
|
|
name='linear2_weight',
|
|
shape=[self.dim_feedforward, self.d_model],
|
|
dtype=self.dtype,
|
|
)
|
|
linear2_bias = paddle.static.data(
|
|
name='linear2_bias', shape=[self.d_model]
|
|
)
|
|
ln1_scale = paddle.static.data(
|
|
name='ln1_scale', shape=[self.d_model]
|
|
)
|
|
ln1_bias = paddle.static.data(name='ln1_bias', shape=[self.d_model])
|
|
ln2_scale = paddle.static.data(
|
|
name='ln2_scale', shape=[self.d_model]
|
|
)
|
|
ln2_bias = paddle.static.data(name='ln2_bias', shape=[self.d_model])
|
|
|
|
fused_out = incubate_f.fused_feedforward(
|
|
x,
|
|
linear1_weight,
|
|
linear2_weight,
|
|
linear1_bias,
|
|
linear2_bias,
|
|
ln1_scale,
|
|
ln1_bias,
|
|
ln2_scale,
|
|
ln2_bias,
|
|
0.0,
|
|
0.0,
|
|
activation="relu",
|
|
pre_layer_norm=False,
|
|
)
|
|
|
|
exe = paddle.static.Executor(get_device_place())
|
|
|
|
fetch = exe.run(
|
|
feed={
|
|
'x': x_data,
|
|
'linear1_weight': linear1_weight_data,
|
|
'linear1_bias': linear1_bias_data,
|
|
'linear2_weight': linear2_weight_data,
|
|
'linear2_bias': linear2_bias_data,
|
|
'ln1_scale': ln1_scale_data,
|
|
'ln1_bias': ln1_bias_data,
|
|
'ln2_scale': ln2_scale_data,
|
|
'ln2_bias': ln2_bias_data,
|
|
},
|
|
fetch_list=[fused_out],
|
|
)
|
|
|
|
return fetch
|
|
|
|
def run_base_ffn(
|
|
self,
|
|
x_data,
|
|
linear1_weight_data,
|
|
linear1_bias_data,
|
|
linear2_weight_data,
|
|
linear2_bias_data,
|
|
ln1_scale_data,
|
|
ln1_bias_data,
|
|
ln2_scale_data,
|
|
ln2_bias_data,
|
|
):
|
|
main = paddle.static.Program()
|
|
startup = paddle.static.Program()
|
|
paddle.seed(42)
|
|
with paddle.static.program_guard(main, startup):
|
|
x = paddle.static.data(
|
|
name='x',
|
|
shape=[self.batch_size, self.d_model, self.dim_feedforward],
|
|
dtype=self.dtype,
|
|
)
|
|
linear1_weight = paddle.static.data(
|
|
name='linear1_weight',
|
|
shape=[self.d_model, self.dim_feedforward],
|
|
dtype=self.dtype,
|
|
)
|
|
linear1_bias = paddle.static.data(
|
|
name='linear1_bias', shape=[self.dim_feedforward]
|
|
)
|
|
linear2_weight = paddle.static.data(
|
|
name='linear2_weight',
|
|
shape=[self.dim_feedforward, self.d_model],
|
|
dtype=self.dtype,
|
|
)
|
|
linear2_bias = paddle.static.data(
|
|
name='linear2_bias', shape=[self.d_model]
|
|
)
|
|
ln1_scale = paddle.static.data(
|
|
name='ln1_scale', shape=[self.d_model]
|
|
)
|
|
ln1_bias = paddle.static.data(name='ln1_bias', shape=[self.d_model])
|
|
ln2_scale = paddle.static.data(
|
|
name='ln2_scale', shape=[self.d_model]
|
|
)
|
|
ln2_bias = paddle.static.data(name='ln2_bias', shape=[self.d_model])
|
|
|
|
# base ffn
|
|
linear1_out = F.linear(x, linear1_weight, linear1_bias)
|
|
act_out = F.relu(linear1_out)
|
|
dropout1_out = F.dropout(x=act_out, p=0.0, training=False)
|
|
linear2_out = F.linear(dropout1_out, linear2_weight, linear2_bias)
|
|
dropout2_out = x + F.dropout(x=linear2_out, p=0.0, training=False)
|
|
ln_out = F.layer_norm(
|
|
dropout2_out,
|
|
normalized_shape=[self.d_model],
|
|
weight=ln2_scale,
|
|
bias=ln2_bias,
|
|
)
|
|
|
|
exe = paddle.static.Executor(get_device_place())
|
|
|
|
fetch = exe.run(
|
|
feed={
|
|
'x': x_data,
|
|
'linear1_weight': linear1_weight_data,
|
|
'linear1_bias': linear1_bias_data,
|
|
'linear2_weight': linear2_weight_data,
|
|
'linear2_bias': linear2_bias_data,
|
|
'ln1_scale': ln1_scale_data,
|
|
'ln1_bias': ln1_bias_data,
|
|
'ln2_scale': ln2_scale_data,
|
|
'ln2_bias': ln2_bias_data,
|
|
},
|
|
fetch_list=[ln_out],
|
|
)
|
|
|
|
return fetch
|
|
|
|
def test_static(self):
|
|
paddle.enable_static()
|
|
|
|
x_data = np.random.random(
|
|
(self.batch_size, self.d_model, self.dim_feedforward)
|
|
).astype(self.dtype)
|
|
linear1_weight_data = np.random.random(
|
|
(self.d_model, self.dim_feedforward)
|
|
).astype(self.dtype)
|
|
linear1_bias_data = np.zeros(self.dim_feedforward).astype(self.dtype)
|
|
linear2_weight_data = np.random.random(
|
|
(self.dim_feedforward, self.d_model)
|
|
).astype(self.dtype)
|
|
linear2_bias_data = np.zeros(self.d_model).astype(self.dtype)
|
|
|
|
ln1_scale_data = np.ones(self.d_model).astype(self.layer_norm_dtype)
|
|
ln1_bias_data = np.zeros(self.d_model).astype(self.layer_norm_dtype)
|
|
ln2_scale_data = np.ones(self.d_model).astype(self.layer_norm_dtype)
|
|
ln2_bias_data = np.zeros(self.d_model).astype(self.layer_norm_dtype)
|
|
|
|
fused_feedforward_res = self.run_fused_feedforward(
|
|
x_data,
|
|
linear1_weight_data,
|
|
linear1_bias_data,
|
|
linear2_weight_data,
|
|
linear2_bias_data,
|
|
ln1_scale_data,
|
|
ln1_bias_data,
|
|
ln2_scale_data,
|
|
ln2_bias_data,
|
|
)
|
|
|
|
base_ffn_res = self.run_base_ffn(
|
|
x_data,
|
|
linear1_weight_data,
|
|
linear1_bias_data,
|
|
linear2_weight_data,
|
|
linear2_bias_data,
|
|
ln1_scale_data,
|
|
ln1_bias_data,
|
|
ln2_scale_data,
|
|
ln2_bias_data,
|
|
)
|
|
|
|
np.testing.assert_allclose(
|
|
fused_feedforward_res, base_ffn_res, rtol=1e-05, atol=0.001
|
|
)
|
|
|
|
|
|
class TestFusedFFNOpError(unittest.TestCase):
|
|
def test_errors(self):
|
|
paddle.enable_static()
|
|
with paddle.static.program_guard(
|
|
paddle.static.Program(), paddle.static.Program()
|
|
):
|
|
|
|
def test_dtype():
|
|
x = paddle.static.data(
|
|
name='x', shape=[1, 10, 10], dtype="int32"
|
|
)
|
|
linear1_weight = paddle.static.data(
|
|
name='linear1_weight', shape=[1, 10, 10], dtype="float32"
|
|
)
|
|
linear2_weight = paddle.static.data(
|
|
name='linear2_weight', shape=[1, 10, 10], dtype="float32"
|
|
)
|
|
incubate_f.fused_feedforward(x, linear1_weight, linear2_weight)
|
|
|
|
self.assertRaises(TypeError, test_dtype)
|
|
|
|
def test_dropout_rate_type():
|
|
x = paddle.static.data(
|
|
name='x1', shape=[1, 10, 10], dtype="float32"
|
|
)
|
|
linear1_weight = paddle.static.data(
|
|
name='linear1_weight1', shape=[10, 10], dtype="float32"
|
|
)
|
|
linear2_weight = paddle.static.data(
|
|
name='linear2_weight1', shape=[10, 10], dtype="float32"
|
|
)
|
|
incubate_f.fused_feedforward(
|
|
x, linear1_weight, linear2_weight, dropout1_rate="a"
|
|
)
|
|
|
|
self.assertRaises(TypeError, test_dropout_rate_type)
|
|
|
|
def test_dropout_rate_value():
|
|
x = paddle.static.data(
|
|
name='x2', shape=[1, 10, 10], dtype="float32"
|
|
)
|
|
linear1_weight = paddle.static.data(
|
|
name='linear1_weight2', shape=[10, 10], dtype="float32"
|
|
)
|
|
linear2_weight = paddle.static.data(
|
|
name='linear2_weight2', shape=[10, 10], dtype="float32"
|
|
)
|
|
incubate_f.fused_feedforward(
|
|
x, linear1_weight, linear2_weight, dropout2_rate=-1
|
|
)
|
|
|
|
self.assertRaises(ValueError, test_dropout_rate_value)
|
|
|
|
def test_dropout_mode():
|
|
x = paddle.static.data(
|
|
name='x3', shape=[1, 10, 10], dtype="float32"
|
|
)
|
|
linear1_weight = paddle.static.data(
|
|
name='linear1_weight3', shape=[10, 10], dtype="float32"
|
|
)
|
|
linear2_weight = paddle.static.data(
|
|
name='linear2_weight3', shape=[10, 10], dtype="float32"
|
|
)
|
|
incubate_f.fused_feedforward(
|
|
x, linear1_weight, linear2_weight, mode='test'
|
|
)
|
|
|
|
self.assertRaises(ValueError, test_dropout_mode)
|
|
|
|
|
|
class APITestStaticFusedFFNZeroSizeTensor(unittest.TestCase):
|
|
def setUp(self):
|
|
self.dtype = "float32"
|
|
self.layer_norm_dtype = "float32"
|
|
self.batch_size = 1
|
|
self.d_model = 8
|
|
self.dim_feedforward = 0
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|