머신러닝

본문 바로가기
사이트 내 전체검색


머신러닝
머신러닝

2. 파이토치 개발환경 구축(Anaconda3)

페이지 정보

작성자 관리자 댓글 0건 조회 1,976회 작성일 20-02-25 12:37

본문

2. 파이토치 개발환경 구축(Anaconda3)

1. 아나콘다3 설치하기


2. 파이썬 환경설정


아나콘다 설치가 끝나면, Anaconda Prompt에서 파이썬 환경 설치, 각종 라이브러리 설치, 스크립트 실행 등의 작업을 한다.


2.1 새로운 파이썬 환경 생성


(base) C:\Users\jklee>conda create -n pytlesson
Collecting package metadata (current_repodata.json): done
Solving environment: done


==> WARNING: A newer version of conda exists. <==
  current version: 4.7.12
  latest version: 4.8.2

Please update conda by running

    $ conda update -n base -c defaults conda


## Package Plan ##

  environment location: C:\Anaconda3\envs\pytlesson


Proceed ([y]/n)? y

 

Preparing transaction: done
Verifying transaction: done
Executing transaction: done
#
# To activate this environment, use
#
#     $ conda activate pytlesson
#
# To deactivate an active environment, use
#
#     $ conda deactivate


(base) C:\Users\jklee>
 

여기서 파이썬 환경 이름을 pytlesson이라고 했지만 다른 이름을 써도 무방하다.


파이썬 환경을 생성하고 나면 아래 명령으로 새론 만든 환경을 활성화 한다.


(base) C:\Users\jklee>activate pytlesson

(pytlesson) C:\Users\jklee>



2.2 파이토치 설치


(pytlesson) C:\Users\jklee>conda install -c peterjc123 pytorch
Collecting package metadata (current_repodata.json): done
Solving environment: done


==> WARNING: A newer version of conda exists. <==
  current version: 4.7.12
  latest version: 4.8.2

Please update conda by running

    $ conda update -n base -c defaults conda


## Package Plan ##

  environment location: C:\Anaconda3\envs\pytlesson

  added / updated specs:
    - pytorch


The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    certifi-2019.11.28         |           py36_0         154 KB
    cffi-1.14.0                |   py36h7a1dbc1_0         222 KB
    intel-openmp-2020.0        |              166         1.5 MB
    mkl-2020.0                 |              166        98.9 MB
    mkl-service-2.3.0          |   py36hb782905_0         210 KB
    mkl_fft-1.0.15             |   py36h14836fe_0         118 KB
    mkl_random-1.1.0           |   py36h675688f_0         233 KB
    numpy-1.18.1               |   py36h93ca92e_0           6 KB
    numpy-base-1.18.1          |   py36hc3f5095_1         3.8 MB
    pip-20.0.2                 |           py36_1         1.7 MB
    pycparser-2.19             |           py36_0         171 KB
    python-3.6.10              |       h9f7ef89_0        15.9 MB
    pytorch-0.3.1              |py36_cuda80_cudnn6he774522_2       349.3 MB  peterjc123
    setuptools-45.2.0          |           py36_0         528 KB
    six-1.14.0                 |           py36_0          27 KB
    sqlite-3.31.1              |       he774522_0         637 KB
    vc-14.1                    |       h21ff451_1           4 KB  peterjc123
    vs2015_runtime-14.16.27012 |       hf0eaf9b_1         1.1 MB
    vs2017_runtime-15.4.27004.2010|                1         1.6 MB  peterjc123
    wheel-0.34.2               |           py36_0          66 KB
    wincertstore-0.2           |   py36h7fe50ca_0          14 KB
    ------------------------------------------------------------
                                           Total:       476.3 MB

The following NEW packages will be INSTALLED:

  blas               pkgs/main/win-64::blas-1.0-mkl
  certifi            pkgs/main/win-64::certifi-2019.11.28-py36_0
  cffi               pkgs/main/win-64::cffi-1.14.0-py36h7a1dbc1_0
  icc_rt             pkgs/main/win-64::icc_rt-2019.0.0-h0cc432a_1
  intel-openmp       pkgs/main/win-64::intel-openmp-2020.0-166
  mkl                pkgs/main/win-64::mkl-2020.0-166
  mkl-service        pkgs/main/win-64::mkl-service-2.3.0-py36hb782905_0
  mkl_fft            pkgs/main/win-64::mkl_fft-1.0.15-py36h14836fe_0
  mkl_random         pkgs/main/win-64::mkl_random-1.1.0-py36h675688f_0
  numpy              pkgs/main/win-64::numpy-1.18.1-py36h93ca92e_0
  numpy-base         pkgs/main/win-64::numpy-base-1.18.1-py36hc3f5095_1
  pip                pkgs/main/win-64::pip-20.0.2-py36_1
  pycparser          pkgs/main/win-64::pycparser-2.19-py36_0
  python             pkgs/main/win-64::python-3.6.10-h9f7ef89_0
  pytorch            peterjc123/win-64::pytorch-0.3.1-py36_cuda80_cudnn6he774522_2
  setuptools         pkgs/main/win-64::setuptools-45.2.0-py36_0
  six                pkgs/main/win-64::six-1.14.0-py36_0
  sqlite             pkgs/main/win-64::sqlite-3.31.1-he774522_0
  vc                 peterjc123/win-64::vc-14.1-h21ff451_1
  vs2015_runtime     pkgs/main/win-64::vs2015_runtime-14.16.27012-hf0eaf9b_1
  vs2017_runtime     peterjc123/win-64::vs2017_runtime-15.4.27004.2010-1
  wheel              pkgs/main/win-64::wheel-0.34.2-py36_0
  wincertstore       pkgs/main/win-64::wincertstore-0.2-py36h7fe50ca_0


Proceed ([y]/n)?y


Downloading and Extracting Packages
mkl-2020.0           | 98.9 MB   | ############################################################################ | 100%
vs2017_runtime-15.4. | 1.6 MB    | ############################################################################ | 100%
python-3.6.10        | 15.9 MB   | ############################################################################ | 100%
sqlite-3.31.1        | 637 KB    | ############################################################################ | 100%
pip-20.0.2           | 1.7 MB    | ############################################################################ | 100%
mkl_fft-1.0.15       | 118 KB    | ############################################################################ | 100%
mkl-service-2.3.0    | 210 KB    | ############################################################################ | 100%
pycparser-2.19       | 171 KB    | ############################################################################ | 100%
certifi-2019.11.28   | 154 KB    | ############################################################################ | 100%
six-1.14.0           | 27 KB     | ############################################################################ | 100%
numpy-1.18.1         | 6 KB      | ############################################################################ | 100%
wincertstore-0.2     | 14 KB     | ############################################################################ | 100%
wheel-0.34.2         | 66 KB     | ############################################################################ | 100%
mkl_random-1.1.0     | 233 KB    | ############################################################################ | 100%
intel-openmp-2020.0  | 1.5 MB    | ############################################################################ | 100%
cffi-1.14.0          | 222 KB    | ############################################################################ | 100%
vc-14.1              | 4 KB      | ############################################################################ | 100%
numpy-base-1.18.1    | 3.8 MB    | ############################################################################ | 100%
setuptools-45.2.0    | 528 KB    | ############################################################################ | 100%
vs2015_runtime-14.16 | 1.1 MB    | ############################################################################ | 100%
pytorch-0.3.1        | 349.3 MB  | ############################################################################ | 100%
Preparing transaction: done
Verifying transaction: done
Executing transaction: done

(pytlesson) C:\Users\jklee>
 

Numpy 등  파이토치를 사용용하는데 필요한 라이브러리도 함께 설치한다.



2.3 실습 관련 라이브러리 설치


jupyter : 개발환경 기능을 제공하는 라이브러리

Pillow : 이미지 처리 관련 함수를 제공

Matplotlib: 그래프 작성용 함수를 제공

Pandas : 데이터를 다루는 함수를 제공

Scikit-learn : 머신러닝 관련 함수를 제공

KoNLPy : 자연어 처리용 함수를 제공


(pytlesson) C:\Users\jklee>conda install jupyter                                                

(pytlesson) C:\Users\jklee>conda install pillow 

(pytlesson) C:\Users\jklee>conda install matplotlib  

(pytlesson) C:\Users\jklee>conda install pandas  

(pytlesson) C:\Users\jklee>conda install scikit-learn 

(pytlesson) C:\Users\jklee>pip install konlpy 



모든 라이브러리가 끝나면 윈도우 탐색기를 열어 C:\Anaconda3\envs\pytlesson 폴더를 확인한다.

envs아래를 보면 방금 추가한 파이썬 환경의 이름인 pytlesson이라는 폴더를 확인할 수 있다.



파이토치 설치

 

conda install pytorch-cpu==1.1.0 torchvision-cpu==0.3.0 cpuonly -c pytorch  

 

댓글목록

등록된 댓글이 없습니다.


개인정보취급방침 서비스이용약관 모바일 버전으로 보기 상단으로

TEL. 063-469-4551 FAX. 063-469-4560 전북 군산시 대학로 558
군산대학교 컴퓨터정보공학과

Copyright © www.leelab.co.kr. All rights reserved.