速度比较

Safetensors 非常快。让我们通过加载 gpt2 的权重来将其与 PyTorch 进行比较。要运行 GPU 基准测试，请确保您的机器有 GPU，或者如果您使用的是 Google Colab，请确保您已选择 GPU 运行时。

在开始之前，请确保您已安装所有必要的库

pip install safetensors huggingface_hub torch

让我们从导入所有将要使用的包开始

>>> import os
>>> import datetime
>>> from huggingface_hub import hf_hub_download
>>> from safetensors.torch import load_file
>>> import torch

下载 gpt2 的 safetensors 和 torch 权重

>>> sf_filename = hf_hub_download("gpt2", filename="model.safetensors")
>>> pt_filename = hf_hub_download("gpt2", filename="pytorch_model.bin")

CPU 基准测试

>>> start_st = datetime.datetime.now()
>>> weights = load_file(sf_filename, device="cpu")
>>> load_time_st = datetime.datetime.now() - start_st
>>> print(f"Loaded safetensors {load_time_st}")

>>> start_pt = datetime.datetime.now()
>>> weights = torch.load(pt_filename, map_location="cpu")
>>> load_time_pt = datetime.datetime.now() - start_pt
>>> print(f"Loaded pytorch {load_time_pt}")

>>> print(f"on CPU, safetensors is faster than pytorch by: {load_time_pt/load_time_st:.1f} X")
Loaded safetensors 0:00:00.004015
Loaded pytorch 0:00:00.307460
on CPU, safetensors is faster than pytorch by: 76.6 X

速度提升是由于该库通过直接映射文件来避免不必要的内存拷贝。这实际上可以在纯 PyTorch 上实现。当前显示的速度提升是在以下环境中获得的：

操作系统：Ubuntu 18.04.6 LTS
CPU：Intel(R) Xeon(R) CPU @ 2.00GHz

GPU 基准测试

>>> # This is required because this feature hasn't been fully verified yet, but 
>>> # it's been tested on many different environments
>>> os.environ["SAFETENSORS_FAST_GPU"] = "1"

>>> # CUDA startup out of the measurement
>>> torch.zeros((2, 2)).cuda()

>>> start_st = datetime.datetime.now()
>>> weights = load_file(sf_filename, device="cuda:0")
>>> load_time_st = datetime.datetime.now() - start_st
>>> print(f"Loaded safetensors {load_time_st}")

>>> start_pt = datetime.datetime.now()
>>> weights = torch.load(pt_filename, map_location="cuda:0")
>>> load_time_pt = datetime.datetime.now() - start_pt
>>> print(f"Loaded pytorch {load_time_pt}")

>>> print(f"on GPU, safetensors is faster than pytorch by: {load_time_pt/load_time_st:.1f} X")
Loaded safetensors 0:00:00.165206
Loaded pytorch 0:00:00.353889
on GPU, safetensors is faster than pytorch by: 2.1 X

速度提升的原因是该库能够跳过不必要的 CPU 内存分配。据我们所知，这在纯 PyTorch 中是无法复现的。该库的工作原理是内存映射文件，使用 PyTorch 创建空张量，然后直接调用 cudaMemcpy 将张量直接移动到 GPU 上。当前显示的速度提升是在以下环境中获得的：

操作系统：Ubuntu 18.04.6 LTS
GPU：Tesla T4
驱动版本：460.32.03
CUDA 版本：11.2

< > 在 GitHub 上更新