wehub-resource-sync
201 lines
7.9 KiB
Python
201 lines
7.9 KiB
Python
# Copyright (c) 2023 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 copy
|
|
|
|
import numpy as np
|
|
import paddle
|
|
import paddle.nn as nn
|
|
|
|
from paddlenlp.transformers import PretrainedConfig, PretrainedModel
|
|
|
|
|
|
def get_pretrain_arguments(pretrain_arguments):
|
|
|
|
configs = {}
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 8
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
configs["TP8"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 2
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding"] = ""
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 4
|
|
configs["TP2"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 4
|
|
train_args["pipeline_parallel_degree"] = 2
|
|
configs["TP4PP2"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 4
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding"] = ""
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 2
|
|
configs["TP4DP2"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 4
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding"] = "stage1"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 2
|
|
configs["TP4Sharding2"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 2
|
|
train_args["pipeline_parallel_degree"] = 4
|
|
configs["TP2PP4"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 2
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding"] = "stage1"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 4
|
|
configs["TP2Sharding4"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 1
|
|
train_args["pipeline_parallel_degree"] = 8
|
|
configs["PP8"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 1
|
|
train_args["pipeline_parallel_degree"] = 4
|
|
train_args["sharding"] = ""
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 2
|
|
configs["PP4DP2"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 1
|
|
train_args["pipeline_parallel_degree"] = 4
|
|
train_args["sharding"] = "stage1"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 2
|
|
configs["PP4Sharding2"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 1
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding"] = "stage1"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 8
|
|
configs["Sharding8S1"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 1
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding"] = "stage2"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 8
|
|
configs["Sharding8S2"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 1
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding_parallel_degree"] = 4
|
|
train_args["sharding"] = "stage1"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 8
|
|
configs["Sharding4S1DP2"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 1
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding_parallel_degree"] = 4
|
|
train_args["sharding"] = "stage2"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 8
|
|
configs["Sharding4S2DP2"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 1
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding_parallel_degree"] = 2
|
|
train_args["sharding"] = "stage1"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 8
|
|
configs["Sharding2S1DP4"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 1
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding_parallel_degree"] = 2
|
|
train_args["sharding"] = "stage2"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 8
|
|
configs["Sharding2S2DP4"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 1
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding_parallel_degree"] = 1
|
|
train_args["sharding"] = "stage2"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 8
|
|
configs["DP8"] = train_args
|
|
|
|
train_args = copy.deepcopy(pretrain_arguments)
|
|
train_args["tensor_parallel_degree"] = 2
|
|
train_args["pipeline_parallel_degree"] = 1
|
|
train_args["sharding_parallel_degree"] = 2
|
|
train_args["sharding"] = "stage1"
|
|
train_args["gradient_accumulation_steps"] = train_args["gradient_accumulation_steps"] // 4
|
|
configs["TP2DP2Sharding2"] = train_args
|
|
|
|
return configs
|
|
|
|
|
|
class RegressionDataset:
|
|
def __init__(self, a=2, b=3, length=64, seed=42, label_names=None):
|
|
np.random.seed(seed)
|
|
self.label_names = ["labels"] if label_names is None else label_names
|
|
self.length = length
|
|
self.x = np.random.normal(size=(length,)).astype(np.float32)
|
|
self.ys = [a * self.x + b + np.random.normal(scale=0.1, size=(length,)) for _ in self.label_names]
|
|
self.ys = [y.astype(np.float32) for y in self.ys]
|
|
|
|
def __len__(self):
|
|
return self.length
|
|
|
|
def __getitem__(self, i):
|
|
result = {name: y[i] for name, y in zip(self.label_names, self.ys)}
|
|
result["input_x"] = self.x[i]
|
|
return result
|
|
|
|
|
|
class RegressionModelConfig(PretrainedConfig):
|
|
def __init__(self, a=0, b=0, double_output=False, random_torch=True, **kwargs):
|
|
super().__init__(**kwargs)
|
|
self.a = a
|
|
self.b = b
|
|
self.double_output = double_output
|
|
self.random_torch = random_torch
|
|
self.hidden_size = 1
|
|
|
|
|
|
class RegressionPretrainedModel(PretrainedModel):
|
|
config_class = RegressionModelConfig
|
|
base_model_prefix = "regression"
|
|
|
|
def __init__(self, config):
|
|
super().__init__(config)
|
|
self.a = paddle.create_parameter(shape=[], dtype=paddle.float32)
|
|
self.b = paddle.create_parameter(shape=[], dtype=paddle.float32)
|
|
self.a.set_value(paddle.to_tensor(config.a, paddle.float32))
|
|
self.b.set_value(paddle.to_tensor(config.b, paddle.float32))
|
|
self.double_output = config.double_output
|
|
|
|
def forward(self, input_x, labels=None, **kwargs):
|
|
y = input_x * self.a + self.b
|
|
if labels is None:
|
|
return (y, y) if self.double_output else (y,)
|
|
loss = nn.functional.mse_loss(y, labels)
|
|
return (loss, y, y) if self.double_output else (loss, y)
|