customer gender customer_ids
1 0 [1,2,3]
2 1 [6,2,1]
3 0 [4,3,9]
我在csv文件中有一些数据如上所述。我像这样加载csv文件
df = pd.read_csv('customer.csv', sep='\t')
这会将customer_ids加载为字符串,如"['1','2','3']", ...
但我需要将customer_ids数据作为像这样的numpy数组
[list([1,2,3]), list([6,2,1]), list([4,3,9])]
答案 0 :(得分:1)
加载数据时,请指定converters
参数 -
df = pd.read_csv('customer.csv', sep='\t', converters={'customer_ids' : pd.eval})
df
customer gender customer_ids
0 1 0 [1, 2, 3]
1 2 1 [6, 2, 1]
2 3 0 [4, 3, 9]
df.customer_ids.tolist()
[[1, 2, 3], [6, 2, 1], [4, 3, 9]]