利用pandas將非數(shù)值數(shù)據(jù)轉(zhuǎn)換成數(shù)值的方式

# 非數(shù)值列處理函數(shù)
def handel_non_numerical_data(df,name): #----------------name是需要處理的列名稱(str)，暫不考慮列表
 nrows = len(df[name])  #----------------數(shù)據(jù)集的行數(shù)
 old_col = df.columns.tolist() #----------------初始的列名集合
 name_index = old_col.index(name) #---------要處理的列的在數(shù)據(jù)集中的索引值
 name_data = df[name].values.tolist()#-----------將要處理煩人列復(fù)制成一個(gè)列表
 df.drop([name],axis =1,inplace =True) 
 unique_kinds = set(name_data)
 convert_dict = {}; x = 0   #構(gòu)造對(duì)應(yīng)種類數(shù)值轉(zhuǎn)化字典
 for i in unique_kinds:
 convert_dict[i] = x
 x += 1
 def convert(val):
 return convert_dict[val] 
 name_data = list(map(convert,name_data))#利用map函數(shù)直接迭代轉(zhuǎn)化
 
 new_col = df.columns.tolist()
 new_col.insert(name_index,name)
 df.reindex(columns = new_col) #----------------重構(gòu)數(shù)據(jù)的列
 df[name] = name_data

def handle_non_numerical_data(df):
 columns = df.columns.values
 for column in columns:
 text_digit_vals = {}
 def convert_to_int(val):
  return text_digit_vals[val]
 if df[column].dtype != np.int64 and df[column].dtype != np.float64:
  column_content = df[column].values.tolist()
  unique_elements = set(column_content)
  print(unique_elements)
  x =0
  for unique in unique_elements:
  if unique not in text_digit_vals:
   text_digit_vals[unique] = x
   x+=1
 df[column] = list(map(convert_to_int,df[column]))