Python read_csv分隔符正在创建无关的空列

我有下面的. txt文件，我想读到一个Pandas Rame中:

#Date                     run               angle                NAME        
#-----                  _______            ________             _______    
2023-02-15 10:00:00      120716            -1.75493           4.5x10-4 Al 40um
2023-02-15 10:38:48      120716            -1.75493           JD70-103 50um 0/90 deg
2023-02-15 18:25:41      120723            -0.658             JD70-103 50um 45/135 deg

我正在try 使用匹配任意数量的空格的正则表达式分隔符读取. txt文件，除非前面有日期("\d\d:\d\d\d")、"Al"、"\d\dum"、"deg"或类似于"[0—999]/[0—999]"的内容:

df = pd.read_csv("file.txt", engine='python', sep='\s+(?!\d\d:\d\d:\d\d|Al|\d\dum|deg|((\d|\d\d|\d\d\d)\/(\d|\d\d|\d\d\d)))')

出于某种原因，这创建了一个将三列NaN值插入到我想要的每一列之间:

                  Date  NaN  None.1  None.2      run   None.3     None.4     None.5        angle ...
0  2023-02-15 10:00:00  NaN     NaN     NaN   120716      NaN        NaN        NaN     -1.75493 ...
1  2023-02-15 10:38:48  NaN     NaN     NaN   120716      NaN        NaN        NaN     -1.75493 ...
2  2023-02-15 18:25:41  NaN     NaN     NaN   120723      NaN        NaN        NaN     -0.658 ...

你知道为什么会这样吗？我最好的猜测是分隔符正在分隔一行中的多个空格，但这不是我期望从上面提到的regex中得到的行为.