Home [Python] EDA 대신 Pandas Profiling
Post
Cancel

[Python] EDA 대신 Pandas Profiling

Example

Default

1
2
3
4
5
6
7
8
9
10
11
12
13
from pandas_profiling import ProfileReport

pr = ProfileReport(pdf2)
pr.to_file(output_file="feature_profiling.html")

# Disable samples, correlations, missing diagrams and duplicates at once
r = ProfileReport(
    samples=None,
    correlations=None,
    missing_diagrams=None,
    duplicates=None,
    interactions=None,
)

Minimal

1
2
3
4
from pandas_profiling import ProfileReport

pr = ProfileReport(pdf2, minimal=True)
pr.to_file(output_file="feature_profiling.html")

Parameters

1
2
3
# Number of workers in thread pool. When set to zero, it is set to the number of CPUs available.
# pool_size, default = 0
pr = ProfileReport(df, minimal=True, pool_size=0) 
  • 기본값으로 모든 thread pool을 사용도록 설정되어 있음

화면에 표시

1
2
# Widget 형태로 보여서 확인하기가 편함 (Tab으로 각각 속성에 대해서 보여줌)
pr.to_widgets()

참고

This post is licensed under CC BY 4.0 by the author.

[Python] Advanced Techniques

[Python] 시계열 예측을 위한 Prophet