Notice
Recent Posts
Recent Comments
Link
일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | ||
6 | 7 | 8 | 9 | 10 | 11 | 12 |
13 | 14 | 15 | 16 | 17 | 18 | 19 |
20 | 21 | 22 | 23 | 24 | 25 | 26 |
27 | 28 | 29 | 30 |
Tags
- IOS
- MySQL
- Excel
- urllib
- ubuntu
- PER
- 함수
- 리눅스
- pandas
- javascript
- 라즈베리파이
- PyQt
- ASP
- 맛집
- 다이어트
- tensorflow
- 날짜
- python
- GIT
- port
- node.js
- 유니티
- mssql
- Unity
- sqlite
- Linux
- swift
- PyQt5
- MS-SQL
- flutter
Archives
아미(아름다운미소)
메모리 최적화 자동화 본문
# 기존 메모리 사용량 확인
df.info(memory_usage='deep')
# 정수형 컬럼 최적화
int_cols = df.select_dtypes(include=['int64']).columns
df[int_cols] = df[int_cols].apply(pd.to_numeric, downcast='integer')
# 실수형 컬럼 최적화
float_cols = df.select_dtypes(include=['float64']).columns
df[float_cols] = df[float_cols].apply(pd.to_numeric, downcast='float')
# 문자열 컬럼은 범주형으로 변환 (고유값이 적은 경우)
obj_cols = df.select_dtypes(include=['object']).columns
for col in obj_cols:
if df[col].nunique() / len(df[col]) < 0.5: # 고유값이 50% 미만일 때
df[col] = df[col].astype('category')
# 최적화 후 메모리 사용량 확인
df.info(memory_usage='deep')
def optimize_memory(df):
# 수치형 데이터 다운캐스트
for col in df.select_dtypes(include=['integer']).columns:
df[col] = pd.to_numeric(df[col], downcast='integer')
for col in df.select_dtypes(include=['float']).columns:
df[col] = pd.to_numeric(df[col], downcast='float')
# 카테고리 변환 (고유값 20개 이하)
for col in df.select_dtypes(include=['object']).columns:
if df[col].nunique() <= 20:
df[col] = df[col].astype('category')
# 불린 타입 변환
for col in df.select_dtypes(include=['bool']).columns:
df[col] = df[col].astype('bool')
# 날짜 변환
for col in df.select_dtypes(include=['object']).columns:
if df[col].head().str.match(r'\d{4}-\d{2}-\d{2}').all():
df[col] = pd.to_datetime(df[col])
return df
# 사용 예시
df = optimize_memory(df)
'랭귀지 > pandas' 카테고리의 다른 글
df비교 (0) | 2025.03.27 |
---|---|
데이터프레임의 메모리 사용량을 최적화하는 함수 (0) | 2025.03.27 |
category로 변환 (0) | 2025.03.27 |
함수 실행 시간 측정 (0) | 2025.03.26 |
merge (0) | 2025.03.26 |
Comments