数据筛选（pandas）

 导出数据中间有若干空格，且有重复数据

关键点1 ：导入数据（pandas）

 df_txt = pd.read_table(filepathNumber,sep='\s+',header=None,names=['PART','Node','Attached elements','U1','U2','U3'])

sep = '\s+'采用正则表达式的方法来判断空格

header = None表示第一行不作为头部 

关键点2：删除重复行

 df_txt = df_txt.drop_duplicates(['PART','Node'],keep='first')#删除重复行

关键点3：pandas按照条件筛选

def condition(x):
    #设置筛选条件
    condition_1_1 = df_txt.PART == 'TANPIAN-1'
    condition_1_2 = df_txt.PART == 'TANPIAN-1-RAD-4'
    condition_1_3 = df_txt.PART == 'TANPIAN-1-RAD-3'
    condition_1_4 = df_txt.PART == 'TANPIAN-1-RAD-2'
    condition_2 = df_txt.Node == 1338
    result = condition_1_1&condition_2 | condition_1_2&condition_2 | condition_1_3&condition_2 | condition_1_4&condition_2
    return result

使用loc筛选

df_txt = df_txt.loc[condition]

关键点4：pandas筛选后并没有按照我所要求的顺序排列

def changeOrder(x):
    # 数据按照指定顺序进行排列
    df_Order = pd.DataFrame({'PART':['TANPIAN-1','TANPIAN-1-RAD-4','TANPIAN-1-RAD-3','TANPIAN-1-RAD-2'],})
    sort_mapping = df_Order.reset_index().set_index("PART")
    x['size_num'] = x['PART'].map(sort_mapping['index'])
    x = x.sort_values('size_num')
    return x

df_txt = changeOrder(df_txt)

关键点5：数据裁切

只引用pandas几行或者几列

df_Number = df_txt[['U1','U2','U3']]

关键点6：每次for循环产生的DateFrame存入到一个新的DateFrame中

df_empty = pd.DataFrame(columns=['U1', 'U2', 'U3'])#定义一个空DateFrame
for filepathNumber in filepath:
    df_empty = df_empty.append(df_Number)

关键点7：使用print不打印index

print(df_Number.to_string(index=False))