cuDF - GPU DataFrames

cuDF - GPU DataFrames

https://github.com/rapidsai/cudf

GitHub - rapidsai/cudf: cuDF - GPU DataFrame Library

 

데이터 로드/조인/집계/필터링 및 기타 데이터 조작을 위한 GPU 기반 데이터 프레임 라이브러리
매우 빠른 C++/CUDA 기반 라이브러리인 libcudf를 이용
pandas 처럼 import 해서 사용 가능
또는 cudf.pandas 를 이용해서 코드 변경 전혀 없이 기존 pandas를 교체하여 GPU 가속 가능

 

 

You can import cudf directly and use it like pandas:

import cudf

tips_df = cudf.read_csv("https://github.com/plotly/datasets/raw/master/tips.csv")
tips_df["tip_percentage"] = tips_df["tip"] / tips_df["total_bill"] * 100

# display average tip by dining party size
print(tips_df.groupby("size").tip_percentage.mean())

 

Or, you can use cuDF as a no-code-change accelerator for pandas, using cudf.pandas. cudf.pandas supports 100% of the pandas API, utilizing cuDF for supported operations and falling back to pandas when needed:

%load_ext cudf.pandas  # pandas operations now use the GPU!

import pandas as pd

tips_df = pd.read_csv("https://github.com/plotly/datasets/raw/master/tips.csv")
tips_df["tip_percentage"] = tips_df["tip"] / tips_df["total_bill"] * 100

# display average tip by dining party size
print(tips_df.groupby("size").tip_percentage.mean())

 

https://rapids.ai/cudf-pandas/

cuDF Pandas