wehub-resource-sync
158 lines
4.8 KiB
Python
158 lines
4.8 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 random
|
|
|
|
import numpy as np
|
|
|
|
import paddle
|
|
import paddle.distributed as dist
|
|
from paddle.io import DataLoader, Dataset
|
|
|
|
os.environ["FLAGS_embedding_deterministic"] = "1"
|
|
os.environ["FLAGS_cudnn_deterministic"] = "1"
|
|
|
|
mesh = dist.ProcessMesh([0, 1], dim_names=["x"])
|
|
data_world_size = mesh.get_dim_size("x")
|
|
dim = 3
|
|
|
|
|
|
def loss_fn(x, label):
|
|
return x
|
|
|
|
|
|
class RandomDataset(Dataset):
|
|
def __init__(self, num_samples=10):
|
|
self.num_samples = num_samples
|
|
|
|
def __getitem__(self, idx):
|
|
inputs = paddle.ones(dim, dtype="float32")
|
|
input_type = paddle.ones([], dtype="int64") * idx % 2
|
|
label = paddle.ones(1, dtype="int64")
|
|
return {"inputs": [inputs, input_type], "label": label}
|
|
|
|
def __len__(self):
|
|
return self.num_samples
|
|
|
|
|
|
class Layer(paddle.nn.Layer):
|
|
def __init__(self):
|
|
super().__init__()
|
|
self.w = self.create_parameter(
|
|
shape=[dim, dim],
|
|
default_initializer=paddle.nn.initializer.Assign(
|
|
0.05 * paddle.ones([dim, dim])
|
|
),
|
|
)
|
|
|
|
def forward(self, x):
|
|
return paddle.matmul(x, self.w)
|
|
|
|
|
|
class DemoModel(paddle.nn.Layer):
|
|
def __init__(self):
|
|
super().__init__()
|
|
self.layer_a = Layer()
|
|
|
|
def forward(self, inputs, input_type):
|
|
tar = paddle.ones_like(input_type) + 3
|
|
inputs = self.layer_a(inputs)
|
|
while not paddle.equal(input_type, tar).all():
|
|
inputs = self.layer_a(inputs)
|
|
input_type = input_type + 1
|
|
|
|
return inputs.mean()
|
|
|
|
|
|
class TestWhileDemo:
|
|
def init_env(self):
|
|
paddle.seed(2024)
|
|
np.random.seed(2024)
|
|
random.seed(2024)
|
|
|
|
def create_data_loader(self):
|
|
dataset = RandomDataset()
|
|
dataloader = DataLoader(
|
|
dataset,
|
|
batch_size=1,
|
|
shuffle=False,
|
|
)
|
|
|
|
dist_dataloader = dist.shard_dataloader(
|
|
dataloader=dataloader,
|
|
meshes=mesh,
|
|
shard_dims="x",
|
|
input_keys=["inputs", "label"],
|
|
is_dataset_splitted=True,
|
|
)
|
|
return dist_dataloader
|
|
|
|
def test_dynamic(self, dist_dataloader):
|
|
dy_layer = DemoModel()
|
|
opt_dy = paddle.optimizer.AdamW(
|
|
learning_rate=0.001, parameters=dy_layer.parameters()
|
|
)
|
|
dist_opt = dist.shard_optimizer(opt_dy)
|
|
dy_loss_list = []
|
|
for step, data in enumerate(dist_dataloader()):
|
|
[inputs, input_type], _ = data["inputs"], data["label"]
|
|
loss = dy_layer(inputs, input_type)
|
|
loss.backward()
|
|
dist_opt.step()
|
|
dist_opt.clear_grad()
|
|
dy_loss_list.append(loss.numpy())
|
|
dy_loss = np.array(dy_loss_list)
|
|
dy_loss = np.mean(dy_loss)
|
|
return dy_loss
|
|
|
|
def test_dynamic2static(self, dist_dataloader):
|
|
paddle.disable_static()
|
|
paddle.base.set_flags({"FLAGS_enable_pir_api": 1})
|
|
dy2static_layer = DemoModel()
|
|
dy2static_opt = paddle.optimizer.AdamW(
|
|
learning_rate=0.001, parameters=dy2static_layer.parameters()
|
|
)
|
|
static_dp_loss_list = []
|
|
dist_model = dist.to_static(
|
|
dy2static_layer, dist_dataloader, loss_fn, dy2static_opt
|
|
)
|
|
dist_model.train()
|
|
for step, data in enumerate(dist_dataloader()):
|
|
loss = dist_model(data["inputs"], data["label"])
|
|
static_dp_loss_list.append(loss)
|
|
dy2static_losses = np.array(static_dp_loss_list)
|
|
pd_partial_loss = paddle.to_tensor(dy2static_losses)
|
|
pd_loss_list = []
|
|
dist.all_gather(pd_loss_list, pd_partial_loss)
|
|
np_dy2static_loss_list = [loss.numpy() for loss in pd_loss_list]
|
|
np_dy2static_loss = np.array(np_dy2static_loss_list)
|
|
np_dy2static_loss = np.mean(np_dy2static_loss)
|
|
return np_dy2static_loss
|
|
|
|
def test_while(self):
|
|
self.init_env()
|
|
dist_dataloader = self.create_data_loader()
|
|
dynamic_loss = self.test_dynamic(dist_dataloader)
|
|
|
|
self.init_env()
|
|
dist_dataloader = self.create_data_loader()
|
|
dy2static_loss = self.test_dynamic2static(dist_dataloader)
|
|
|
|
np.testing.assert_allclose(dynamic_loss, dy2static_loss, atol=1e-8)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
TestWhileDemo().test_while()
|