Question

我希望通过特征分组将设备日期范围和特征的数据框转换为年度总安装时间。我想翻译一个像这样的数据框：

df_eq=pd.DataFrame({'equip':np.arange(0,10),'char1':[4]*4+[1,2,3]+[5]*3,
                    'char2':['A']*3+['B']*3+['C']*4,
                    'start':pd.to_datetime(['2010-01-10', '2010-01-10','2011-02-24','2011-06-06','2013-09-30','2010-01-10', '2010-01-10','2011-02-24','2011-06-06','2013-09-30']),
                    'end':pd.to_datetime(['2014-05-05']*2+['2015-01-01']*3+[None]*5)})

df_eq
   char1 char2        end  equip      start
0      4     A 2014-05-05      0 2010-01-10
1      4     A 2014-05-05      1 2010-01-10
2      4     A 2015-01-01      2 2011-02-24
3      4     B 2015-01-01      3 2011-06-06
4      1     B 2015-01-01      4 2013-09-30
5      2     B        NaT      5 2010-01-10
6      3     C        NaT      6 2010-01-10
7      5     C        NaT      7 2011-02-24
8      5     C        NaT      8 2011-06-06
9      5     C        NaT      9 2013-09-30

NaT的{{1}}日期时间代表尚未退休的设备。使用这个数据框我想翻译产生以下样本，其中数量是给定年份内单位的安装时间：

end

我可以使用以下代码生成所需的表，但是我想看看是否有更多的pythonic方式使用pandas来生成相同的输出表：

   char1      2011  2012      2013      2014
0      1  0.000000     0  0.254795  1.000000
1      2  1.000000     1  1.000000  1.000000
2      3  1.000000     1  1.000000  1.000000
3      4  3.424658     4  4.000000  2.684932
4      5  1.424658     2  2.254795  3.000000

   char1 char2      2011  2012      2013      2014
0      1     B  0.000000     0  0.254795  1.000000
1      2     B  1.000000     1  1.000000  1.000000
2      3     C  1.000000     1  1.000000  1.000000
3      4     A  2.852055     3  3.000000  1.684932
4      4     B  0.572603     1  1.000000  1.000000
5      5     C  1.424658     2  2.254795  3.000000

Answer 1

我认为您可以使用private void LoadDynamincVideoViews(final String path, String ID) { final float scale = getApplicationContext().getResources() .getDisplayMetrics().density; int pixels = (int) (200 * scale + 0.5f); int pixels1 = (int) (250 * scale + 0.5f); int height = (Devicewidth * 9 / 16); mVideoView = new VideoView(this); mVideoView.setTag(ID); LinearLayout.LayoutParams params = new LinearLayout.LayoutParams( pixels, height); params.setMargins(10, 5, 5, 5); mVideoView.setLayoutParams(params); mVideoViewArray.put(ID, mVideoView); mLinearBottom.addView(mVideoView); // HashMap< String, VideoView> hashmap = new HashMap<String, // VideoView>(); Handler handler = new Handler(); handler.postDelayed(new Runnable() { @Override public void run() { // TODO Auto-generated method stub MediaController mediaController = new MediaController( getApplicationContext()); mediaController.setAnchorView(mVideoView); // mediaController.setMediaPlayer(videoview); Uri video = Uri.parse(path); mVideoView.setMediaController(mediaController); mVideoView.setVideoURI(video); mVideoView.start(); } }, 2000); }向您的部分代码进行矢量化，如下所示：

.where

基于设备日期范围的熊猫高效计算年库存量的方法

1 个答案: