包含查找Pandas

Question

请帮助，我需要创建一个新列。如果我的列类型（列表）包含值。你知道吗

如果类型包含：lookuptable['first']和lookuptable['second']

从查找表返回相应的值。 我尝试了很多方法，对python还是陌生的，希望我能接近

Return:lookuptable['mid-genre']

dh是下面的数据帧

sub-genre   first   second  mid-genre   genre
indie       indie           Alternative rock
dream pop   dream   pop     Alternative rock
shoegaze    shoegaze        Alternative rock
post-hardcore post hardcore HardcorePunk rock
emo          emo            HardcorePunk rock
screamo     screamo         HardcorePunk rock
synthcore   synthcore       Harcore Punk rock 
rock        rock            Contemporary rock

diy=下面的数据帧

artist  genres                             New Column
 2:54   ['metropopolis']                    No Genre (blank)
 22     ['norwegian rock']                  Contemporary
 27     ['boston rock']                     Contemporary
 33     []                                  No Genre (blank)
 36     ['ambient', 'compositional ambient', 'drift', ...
 44     ['emo', 'pop punk', 'skate punk']   Hardcore Punk
 52     []
 68     []
 83     ['hip hop quebecois']               Hip hop

下面的代码尝试

diy = pd.DataFrame(data[['artist','genres']])

for i in diy['genres'].iteritems():
    for x, y, z, t in zip(dh['first'], dh['second'],dh['mid-genre'],dh['genre']):
        if h.str.contains(x) and h.str.contains(z):
            diy['mid-genre'] = z
            diy['Main-genre'] = t

错误消息

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

第二次尝试时，我在IF语句中添加了.any（），以尝试处理异常：

if h.str.contains(x).any() and h.str.contains(z).any():
UserWarning: This pattern has match groups. To actually get the groups, use str.extract.

Answer 1

我想出了解决办法，以防其他人有类似的任务。你知道吗

import re
diy = pd.DataFrame(data[['artist','genres']])

omg = []  # Create container for new values

# Zip lookup table to Loop through strings (Pattern to look for)
for x, y, z, t in zip(lookup['first'], lookup['second'],lookup['mid-genre'],lookup['genre']):
    # convert X and Y to Regular expression pattern
    p = re.compile(x)   
    q = re.compile(str(y)) 

    # Loop through Data    
    for i, k in zip(diy['artist'],diy['genres']):

        # Create and store match object (RegEx object)
        m = p.search(str(k))
        j = q.search(str(k))

        # If M and J both match diy['genre']
        if (m and j):
            woo = (i, z, t)   # Return lookup[['mid-genre','Main-genre']]
            omg.append(woo)   # Append to container
        else:
            # If no match label No genre
            woo = (i,'No genre','No genre') 
            omg.append(woo)