Question

您好我已经很长时间R用户并且慢慢转向Python。

我在R中有一些split＆amp; apply＆amp; combine例程但是我很难找到python相当于这些函数，如grepl，paste，select等。我想要做的是一步一步

读取文件夹
使用“TRIAL”字符串
选择一些特定列并命名此列
当第二列值0.1重新出现时，在新列“否”中创建组序列
将第二列乘以2并创建一个新列
将文件名添加为每个文件的新列

最后制作一个data.frame输出。

         tt      m1      m2      m3      m4      m5      m6  No  tt2 file_name
1       0.10 -0.0047 -0.0168 -0.9938 -0.0087 -0.0105 -0.9709   1  0.2   sum_W_1
2       0.20 -0.0121  0.0002 -0.9898 -0.0364 -0.0027 -0.9925   1  0.4   sum_W_1
3       0.30  0.0193 -0.0068 -0.9884  0.0040  0.0139 -0.9782   1  0.6   sum_W_1
4       0.40 -0.0157  0.0183 -0.9879 -0.0315 -0.0311 -0.9908   1  0.8   sum_W_1
5       0.50 -0.0402  0.0300 -0.9832 -0.0093  0.0269 -0.9781   1  1.0   sum_W_1

这是示例[file] [2]

头（sum_data）

TRIAL :            1        3331        9091
  TRIAL :            2  1384786531   278055555
    2     0.10     0.000E+00 -0.0047 -0.0168 -0.9938    -0.0087 -0.0105 -0.9709     0.0035  0.0079 -0.9754     0.0081  0.0023  0.9997      -0.135324E-09    0.278754E-01
    2     0.20     0.000E+00 -0.0121  0.0002 -0.9898    -0.0364 -0.0027 -0.9925    -0.0242 -0.0050 -0.9929     0.0029 -0.0023  0.9998      -0.133521E-09    0.425567E-01
    2     0.30     0.000E+00  0.0193 -0.0068 -0.9884     0.0040  0.0139 -0.9782    -0.0158  0.0150 -0.9814     0.0054 -0.0008  0.9997      -0.134103E-09    0.255356E-01
    2     0.40     0.000E+00 -0.0157  0.0183 -0.9879    -0.0315 -0.0311 -0.9908    -0.0314 -0.0160 -0.9929     0.0040  0.0010  0.9998      -0.134819E-09    0.257300E-01
    2     0.50     0.000E+00 -0.0402  0.0300 -0.9832    -0.0093  0.0269 -0.9781    -0.0326  0.0247 -0.9802     0.0044 -0.0010  0.9997      -0.131515E-09    0.440350E-01

我尝试了什么，

import os
import glob # damla, topak
import pandas as pd
import numpy

filelist=glob.glob('*.txt')
print(filelist)

names_cols=['tt','m1','m2','m3','m4','m5','m6','m7']

for file in filelist:
    df=pd.read_table(file, header=None,skiprows=7,skipfooter=0,names=names_cols,usecols=[1,5,6,7,8,9,10,11])
    df_del=df[df.V1.str.contains('TRIAL')==False]
    concatdf=pd.concat(df_del,axis=0)

CParserError：指定的列太多：预期为7，找到1

即使这个简单的阶段我也卡住了！

请帮助完成此脚本！

Answer 1

考虑以下翻译。而且由于R＆＃39; lapply将数据帧保存到列表中，您需要在Python中将对应的数据帧附加到初始化列表中：

names_cols = ['tt','m1','m2','m3','m4','m5','m6','m7']
v1 = ["W_1","B_1"]

dfs = []
for file in filelist:
    df = pd.read_table(file, header=None, skiprows=7, skipfooter=0, sep="\s+",
                       names=names_cols, usecols=[1,5,6,7,8,9,10,11])

    df = df[~df['tt'].str.contains('TRIAL|:')]        # KEEP ROWS WITHOUT TRIAL AND COLON
    df['tt'] = df['tt'].astype(float)                 # CONVERT TO FLOAT COLUMN
    df['tt2'] = df['m2'] * 2                          # MULTIPLY BY 2 (DOES NOT CHANGE SIGN)
    df['No'] = (df['tt'].cumsum()==0.1).astype(int)   # BOOLEAN OF A SERIES CUMSUM()        
    df['file_name'] = file[0:3]+'_'+ v1[0]            # EXTRACT FIRST THREE LETTERS
    dfs.append(df)                                    # APPEND TO LIST

print(dfs[0].head())
#     tt      m1      m2      m3      m4      m5      m6      m7     tt2  No file_name
# 0  0.6 -0.9872  0.0119 -0.0119 -0.9883  0.0306 -0.0259 -0.9903  0.0238   0   Spl_W_1
# 1  0.7 -0.9877 -0.0382 -0.0227 -0.9803 -0.0293 -0.0252 -0.9864 -0.0764   0   Spl_W_1
# 2  0.8 -0.9859 -0.0256  0.0218 -0.9829 -0.0323 -0.0098 -0.9870 -0.0512   0   Spl_W_1
# 3  0.9 -0.9838 -0.0030 -0.0032 -0.9844  0.0048 -0.0206 -0.9866 -0.0060   0   Spl_W_1
# 4  1.0 -0.9885 -0.0346 -0.0061 -0.9865 -0.0259 -0.0105 -0.9887 -0.0692   0   Spl_W_1

对于较短的线条，熊猫＆＃39; assign可以最好地作为R dplyr mutate的副本，具有多列分配，但请注意多个分配的列按字母顺序排序：

for file in filelist:
    df = pd.read_table(file, header=None, skiprows=7, skipfooter=0, sep="\s+",
                       names=names_cols, usecols=[1,5,6,7,8,9,10,11])

    df = df[~df['tt'].str.contains('TRIAL|:')].assign(tt2 = df['m2'] * 2, file_name = file[0:3]+'_'+ v1[0])
    df = df.assign(tt = df['tt'].astype(float), No = (df['tt'].astype(float).cumsum()==0.1).astype(int))
    dfs.append(df)

顺便说一句，为什么慢慢从R转移到Python？使用两种很棒的语言！

R's split＆amp; apply＆amp;在Python中结合例程

1 个答案: