将pandas数据帧转换为parquet格式并上传到s3存储桶

import pandas as pd
import pyarrow as pa
import s3fs

data = {'col1': [1, 2, 3, 4, 5],
        'col2': ['a', 'b', 'c', 'd', 'e']}
df = pd.DataFrame(data)

table = pa.Table.from_pandas(df)

s3 = s3fs.S3FileSystem(anon=False)  # 需要提供AWS的访问密钥和密钥ID

bucket_name = 'your_bucket_name'
file_path = 'path/to/your/file.parquet'

with s3.open(f'{bucket_name}/{file_path}', 'wb') as f:
    pa.parquet.write_table(table, f)