使用 PyTorch 训练编译器运行 SageMaker 训练作业

要编译和训练 PyTorch 模型，请使用 SageMaker 训练编译器配置 A SageMaker I PyTorch 估算器，如以下代码示例所示。

from sagemaker.pytorch import PyTorch, TrainingCompilerConfig

# the original max batch size that can fit into GPU memory without compiler
batch_size_native=12
learning_rate_native=float('5e-5')

# an updated max batch size that can fit into GPU memory with compiler
batch_size=64

# update learning rate
learning_rate=learning_rate_native/batch_size_native*batch_size

hyperparameters={
    "n_gpus": 1,
    "batch_size": batch_size,
    "learning_rate": learning_rate
}

pytorch_estimator=PyTorch(
    entry_point='train.py',
    source_dir='path-to-requirements-file', # Optional. Add this if need to install additional packages.
    instance_count=1,
    instance_type='ml.p3.2xlarge',
    framework_version='1.13.1',
    py_version='py3',
    hyperparameters=hyperparameters,
    compiler_config=TrainingCompilerConfig(),
    disable_profiler=True,
    debugger_hook_config=False
)

pytorch_estimator.fit()

要使用编译和训练转换器模型 PyTorch，请使用训练编译器配置 A SageMaker I Hugging Face 估计器 SageMaker ，如以下代码示例所示。

from sagemaker.huggingface import HuggingFace, TrainingCompilerConfig

# the original max batch size that can fit into GPU memory without compiler
batch_size_native=12
learning_rate_native=float('5e-5')

# an updated max batch size that can fit into GPU memory with compiler
batch_size=64

# update learning rate
learning_rate=learning_rate_native/batch_size_native*batch_size

hyperparameters={
    "n_gpus": 1,
    "batch_size": batch_size,
    "learning_rate": learning_rate
}

pytorch_huggingface_estimator=HuggingFace(
    entry_point='train.py',
    instance_count=1,
    instance_type='ml.p3.2xlarge',
    transformers_version='4.21.1',
    pytorch_version='1.11.0',
    hyperparameters=hyperparameters,
    compiler_config=TrainingCompilerConfig(),
    disable_profiler=True,
    debugger_hook_config=False
)

pytorch_huggingface_estimator.fit()

要准备训练脚本，请参阅以下页面。

要查找 end-to-end示例，请参阅以下笔记本：

对于 PyTorch v1.12，您可以通过在 SageMaker AI PyTorch 估算器类的distribution参数中添加指定的pytorch_xla选项，使用 SageMaker 训练编译器运行分布式训练。

from sagemaker.pytorch import PyTorch, TrainingCompilerConfig

# choose an instance type, specify the number of instances you want to use,
# and set the num_gpus variable the number of GPUs per instance.
instance_count=1
instance_type='ml.p3.8xlarge'
num_gpus=4

# the original max batch size that can fit to GPU memory without compiler
batch_size_native=16
learning_rate_native=float('5e-5')

# an updated max batch size that can fit to GPU memory with compiler
batch_size=26

# update learning rate
learning_rate=learning_rate_native/batch_size_native*batch_size*num_gpus*instance_count

hyperparameters={
    "n_gpus": num_gpus,
    "batch_size": batch_size,
    "learning_rate": learning_rate
}

pytorch_estimator=PyTorch(
    entry_point='your_training_script.py',
    source_dir='path-to-requirements-file', # Optional. Add this if need to install additional packages.
    instance_count=instance_count,
    instance_type=instance_type,
    framework_version='1.13.1',
    py_version='py3',
    hyperparameters=hyperparameters,
    compiler_config=TrainingCompilerConfig(),
    distribution ={'pytorchxla' : { 'enabled': True }},
    disable_profiler=True,
    debugger_hook_config=False
)

pytorch_estimator.fit()

PyTorch 在 v1.11 及更高版本中， SageMaker 训练编译器可用于分布式训练，参数中指定了pytorch_xladistribution选项。

from sagemaker.huggingface import HuggingFace, TrainingCompilerConfig

# choose an instance type, specify the number of instances you want to use,
# and set the num_gpus variable the number of GPUs per instance.
instance_count=1
instance_type='ml.p3.8xlarge'
num_gpus=4

# the original max batch size that can fit to GPU memory without compiler
batch_size_native=16
learning_rate_native=float('5e-5')

# an updated max batch size that can fit to GPU memory with compiler
batch_size=26

# update learning rate
learning_rate=learning_rate_native/batch_size_native*batch_size*num_gpus*instance_count

hyperparameters={
    "n_gpus": num_gpus,
    "batch_size": batch_size,
    "learning_rate": learning_rate
}

pytorch_huggingface_estimator=HuggingFace(
    entry_point='your_training_script.py',
    instance_count=instance_count,
    instance_type=instance_type,
    transformers_version='4.21.1',
    pytorch_version='1.11.0',
    hyperparameters=hyperparameters,
    compiler_config=TrainingCompilerConfig(),
    distribution ={'pytorchxla' : { 'enabled': True }},
    disable_profiler=True,
    debugger_hook_config=False
)

pytorch_huggingface_estimator.fit()

对于支持的 PyTorch v1.10.2 及更早版本，Training Compiler 需要另一种机制来启动分布式 SageMaker 训练作业。要运行分布式训 SageMaker 练，Training Compiler 要求您将 SageMaker AI 分布式训练启动器脚本传递给entry_point参数，并将训练脚本传递给hyperparameters参数。以下代码示例显示了如何配置应用所需更改的 A SageMaker I Hugging Face 估算器。

from sagemaker.huggingface import HuggingFace, TrainingCompilerConfig

# choose an instance type, specify the number of instances you want to use,
# and set the num_gpus variable the number of GPUs per instance.
instance_count=1
instance_type='ml.p3.8xlarge'
num_gpus=4

# the original max batch size that can fit to GPU memory without compiler
batch_size_native=16
learning_rate_native=float('5e-5')

# an updated max batch size that can fit to GPU memory with compiler
batch_size=26

# update learning rate
learning_rate=learning_rate_native/batch_size_native*batch_size*num_gpus*instance_count

training_script="your_training_script.py"

hyperparameters={
    "n_gpus": num_gpus,
    "batch_size": batch_size,
    "learning_rate": learning_rate,
    "training_script": training_script     # Specify the file name of your training script.
}

pytorch_huggingface_estimator=HuggingFace(
    entry_point='distributed_training_launcher.py',    # Specify the distributed training launcher script.
    instance_count=instance_count,
    instance_type=instance_type,
    transformers_version='4.17.0',
    pytorch_version='1.10.2',
    hyperparameters=hyperparameters,
    compiler_config=TrainingCompilerConfig(),
    disable_profiler=True,
    debugger_hook_config=False
)

pytorch_huggingface_estimator.fit()

启动器脚本应如下所示。它包装训练脚本，并根据所选训练实例的大小配置分布式训练环境。

# distributed_training_launcher.py

#!/bin/python

import subprocess
import sys

if __name__ == "__main__":
    arguments_command = " ".join([arg for arg in sys.argv[1:]])
    """
    The following line takes care of setting up an inter-node communication
    as well as managing intra-node workers for each GPU.
    """
    subprocess.check_call("python -m torch_xla.distributed.sm_dist " + arguments_command, shell=True)