PlaidML은 nvidia GPU를 사용할 수 없을때 deep learning neural network를 빠르게 훈련시키고 추론하는데 사용하는 라이브러리이다. 최근 뉴스로는 macbook m1 chip에서 pytorch, tensorflow등이 GPU로 바로 작동한다고 하는데, 그 이전에 AMD GPU나 Intel GPU등에서 사용할 수 있는 라이브러리이다. 단점으로는 예전 버전의 keras에 연동되어 있는데, 기본적인 공부를 하는데에는 무리없이 사용할 수 있을 것 같다. keras가 처음 머신러닝이 입문하는데 가장 편한 라이브러리 인 것 같다.

Installation

> pyenv global 2.7.18 3.8.10
> pip3 install virtualenv
> virtualenv plaidml
> source plaidml/bin/activate
> pip3 install plaidml-keras plaidbench
> plaidml-setup

There is an error on python 3.8.10 and you can fix that following the instruction in https://github.com/plaidml/plaidml/issues/1705

> nvim .../lib/python3.9/site-packages/keras/engine/saving.py
fix line 1004, 1008 to comment out decode

If you try on python 3.10.4 and you will find this error https://github.com/plaidml/plaidml/issues/1966

It is because plaidml was developed on old version of keras and h5py

Benchmarking

Using llvm_cpu.0

> plaidbench keras mobilenet

Running 1024 examples with mobilenet, batch size 1, on backend plaid
INFO:plaidml:Opening device "llvm_cpu.0"
Compiling network... Warming up... Running...
Example finished, elapsed: 3.536s (compile), 126.724s (execution)

-----------------------------------------------------------------------------------------
Network Name         Inference Latency         Time / FPS
-----------------------------------------------------------------------------------------
mobilenet            123.75 ms                 120.58 ms / 8.29 fps
Correctness: PASS, max_error: 1.7640328223933466e-05, max_abs_error: 7.040798664093018e-07, fail_ratio: 0.0

Using metal_intel_iris_pro_graphics.0

> plaidbench keras mobilenet

Running 1024 examples with mobilenet, batch size 1, on backend plaid
INFO:plaidml:Opening device "metal_intel_iris_pro_graphics.0"
Compiling network... Warming up... Running...
Example finished, elapsed: 1.878s (compile), 79.834s (execution)

-----------------------------------------------------------------------------------------
Network Name         Inference Latency         Time / FPS
-----------------------------------------------------------------------------------------
mobilenet            77.96 ms                  0.00 ms / 1000000000.00 fps
Correctness: PASS, max_error: 7.6248343248153105e-06, max_abs_error: 2.8312206268310547e-07, fail_ratio: 0.0