"Tensorflow gpu를 사용하면서 필요한 CUDA, cuDNN 설치하기"
1. pip freeze
아나콘다를 통해 깔려있는 모든 것들 나열
2. 자신의 환경과 맞는 CUDA, cuDNN 확인하기
www.tensorflow.org/install/source_windows#tested_build_configurations
cuDDN: 7
CUDA: 9
3. CUDA 설치
https://developer.nvidia.com/cuda-90-download-archive
CUDA Toolkit 9.0 Downloads
Select Target Platform Click on the green buttons that describe your target platform. Only supported platforms will be shown. Operating System Architecture Distribution Version Installer Type Do you want to cross-compile? Yes No Select Host Platform Click
developer.nvidia.com
4. CUDA와 호환되는 버전의 cuDNN 설치
ttps://developer.nvidia.com/rdp/cudnn-archive
5. cuDNN 파일의 압축풀기, C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v9.0 의 폴더로 옮기기
CUDA 버전이 9.0이면 파일의 옮길 경로 = C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v9.0
압축을 푼 cuDNN 파일을 C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v9.0 로 옮겨서 bin, include, lib 폴더를 덮어씌우기
+ 3개의 파일 경로 환경변수 PATH에 추가하기
참고
https://coding-groot.tistory.com/87
error
ImportError: Could not find 'cudart64_90.dll'. TensorFlow requires that this DLL be installed in a directory that is named in your %PATH% environment variable. Download and install CUDA 9.0 from this URL: https://developer.nvidia.com/cuda-90-download-archive
해결방안
https://developer.nvidia.com/cuda-90-download-archive
CUDA Toolkit 9.0 Downloads
Select Target Platform Click on the green buttons that describe your target platform. Only supported platforms will be shown. Operating System Architecture Distribution Version Installer Type Do you want to cross-compile? Yes No Select Host Platform Click
developer.nvidia.com
tensorflow - keras version 매칭
docs.floydhub.com/guides/environments/
List of Available Environments - FloydHub Documentation
edit Environments Below is the list of Deep Learning environments supported by FloydHub. Any of these can be specified in the floyd run command using the --env option. If no --env is provided, it uses the tensorflow-1.9 image by default, which comes with P
docs.floydhub.com
'AI > deep learning' 카테고리의 다른 글
| [Microsoft Malware Prediction] Model Train (0) | 2020.06.15 |
|---|---|
| [Microsoft Malware Prediction] Data analysis (0) | 2020.05.18 |
| Object Detection: Faster-RCNN Model (0) | 2020.05.07 |
| [DeepLearning] Tensorflow Object Detection을 통한 실시간 객체 탐지 (1513) | 2020.04.02 |
| [Tensorflow] 아나콘다(Anaconda) 설치하기 (2) | 2020.04.02 |