[問題] mxnet安裝問題

[jasonfghx (工人)]

[問題類型]:
程式功能諮詢(我想用R 做某件事情，但是我不知道要怎麼用R 寫出來)

[軟體熟悉度]:
入門(寫過其他程式，只是對語法不熟悉)

[問題敘述]:

不好意思

我想在R安裝mxnet  但似乎都不太可行  好向該套件只支援python C++

請問是否有什麼方法可以安裝在R studio上

謝謝回答

[程式範例]:


[關鍵字]:

--
※ 發信站: 批踢踢實業坊(ptt.cc), 來自: 140.116.234.34
※ 文章網址: https://www.ptt.cc/bbs/R_Language/M.1528796341.A.7EA.html
→ andrew43: 試了一下官網的操作但回傳了404 Not Found
→ andrew43: 不過它竟然是要mxnet_0.10.1.tar.gz，這版本應該很舊了
→ andrew43: 目前已經1.2.0版了。可能支持R有得等喔。
→ andrew43: 或許試試直接從github裝source看看。
→ jasonfghx: 這就是我最近在努力的 只是不太會用
→ jasonfghx: 我不太知道如何從github裝source
→ andrew43: 不知從哪版開始就沒有R source了，可能只能裝舊版
→ celestialgod: 下載VS 2017 community，自己編譯好libmxnet
→ andrew43: 我把0.10的source下載並本地端安裝，沒有編譯成功 sorry
→ celestialgod: 我的文章：https://goo.gl/rGXa7o

a<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/callback.R')

b<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/context.R')

c<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/executor.R')

d<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/initializer.R')

e<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/io.R')

f<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/kvstore.R')

g<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/lr_scheduler.R')

h<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/metric.R')

i<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/mlp.R')

j<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/model.R')

k<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/model.rnn.R')

l<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/mx.io.bucket.iter.R')

m<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/ndarray.R')

n<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/optimizer.R')

o<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/profiler.R')

p<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/random.R')

q<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/rnn.graph.R')

r<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/rnn.infer.R')

s<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/symbol.R')

t<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/util.R')

u<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/viz.graph.R')

v<-source('https://raw.githubusercontent.com/apache/incubator-mxnet/master/
R-package/R/zzz.R')

https://github.com/apache/incubator-mxnet/tree/master/R-package/R

請問一下我是這樣把所有的內容load進去就等於載入mxnet?

謝謝
※ 編輯: jasonfghx (223.139.253.5), 06/12/2018 19:16:09
→ celestialgod: mxnet還有一些C++檔案去link libmxnet
→ celestialgod: 這樣是沒用的
→ jasonfghx: 謝謝回復 吾初學者  先從簡單的開始
推 JustInn: 是想用 r跑nn嗎？
推 ching0629: 初學者先裝cpu版本吧，這個套件很好用，我自己覺得自
→ ching0629: 由度遠遠超tenserflow,但缺少文檔很麻煩。
推 ching0629: 另外，如果想裝gpu版本，我覺得直接把Nvidia的NCG Dock
→ ching0629: er image裝下來最快，直接一行指令就搞定運行環境