AttributeError:将Pandas数据框写入S3时，“”DataFrame“”对象没有属性“”_mgr“”

Question

在将pandas数据帧写入S3时获取AttributeError: 'DataFrame' object has no attribute '_mgr'。

import awswrangler as wr
window = '0112'
wr.s3.to_csv(
    df=mergeDf,
    path="s3://MYBUCKET/wrangled/" + "WiderData_2" +window+ ".csv"
)

错误：

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-65-30cb4c51bd25> in <module>
      7 # Use 'w' for py3, 'wb' for py2
      8 with s3.open('........','w') as f:
----> 9     mergeDf.to_csv(f)
.....
~/anaconda3/envs/python3/lib/python3.6/site-packages/pandas/core/generic.py in __getattr__(self, name)
   5272         """
   5273         Return a Numpy representation of the DataFrame.
-> 5274 
   5275         .. warning::
   5276 

AttributeError: 'DataFrame' object has no attribute '_mgr'

也尝试过

import s3fs
s3 = s3fs.S3FileSystem(anon=False)
# Use 'w' for py3, 'wb' for py2
with s3.open('MYBUCEKT/wrangled/WiderData_1011.csv','w') as f:
    mergeDf.to_csv(f)

相同的错误'AttributeError: 'DataFrame' object has no attribute '_mgr' '

Answer 1

只需使用boto3即可

    import boto3
    import io
    import pandas as pd
    bucket='yourbucket'
    key='path/to/file.csv'
    csv_buffer = io.StringIO()
    df.to_csv(csv_buffer)
    response = boto3.client('s3').put_object(Body=csv_buffer.getvalue(), Bucket=bucket, Key=key)