python - 使用 block 大小将 CSV 文件从 s3 加载到 Pandas

Question

我正在尝试使用...从 s3 读取一个非常大的文件

import pandas as pd
import s3fs
df = pd.read_csv('s3://bucket-name/filename', chunksize=100000)

但即使在给出块大小之后，它也会永远占用。是否 chunksize从 s3 获取文件时的选项工作？如果没有，从 s3 加载大文件有什么更好的方法吗？

Answer 1

https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.read_csv.html
明确地说

filepath_or_bufferstr, path object or file-like object Any validstring path is acceptable. The string could be a URL. Valid URLschemes include http, ftp, s3, gs, and file. For file URLs, a host isexpected. A local file could be: file://localhost/path/to/table.csv.

If you want to pass in a path object, pandas accepts any os.PathLike.

By file-like object, we refer to objects with a read() method, such asa file handle (e.g. via builtin open function) or StringIO.

在块中读取时，pandas 会返回您的迭代器对象，您需要对其进行迭代。
就像是:

for df in pd.read_csv('s3://<<bucket-name>>/<<filename>>',chunksize = 100000):
    process df chunk..

如果你认为这是因为 chunksize 很大，你可以考虑为第一个 chunk 尝试它，只为一个小的 chunksize 像这样:

for df in pd.read_csv('s3://<<bucket-name>>/<<filename>>',chunksize = 1000):
    print(df.head())
    break