GroupBy والتجميع والجداول المحورية

أساسيات groupby

import pandas as pd

df = pd.DataFrame({
    "dept":["HR","IT","HR","IT","HR"],
    "salary":[60,80,65,90,70]
})
print(df.groupby("dept")["salary"].mean())
# HR    65.0
# IT    85.0

دوال التجميع المتعددة

import pandas as pd

df = pd.DataFrame({"dept":["HR","IT","HR","IT"],"sal":[60,80,65,90]})
result = df.groupby("dept")["sal"].agg(["mean","max","count"])
print(result)

التجميعات المُسمّاة

import pandas as pd

df = pd.DataFrame({"dept":["HR","IT","HR","IT"],"sal":[60,80,65,90]})
result = df.groupby("dept").agg(
    avg_sal=("sal","mean"),
    max_sal=("sal","max"),
    count=("sal","size")
)
print(result)

استخدام groupby مع أعمدة متعددة

import pandas as pd

df = pd.DataFrame({
    "year":[2023,2023,2024,2024],
    "dept":["HR","IT","HR","IT"],
    "sal":[60,80,65,90]
})
print(df.groupby(["year","dept"])["sal"].sum())

transform()

import pandas as pd

df = pd.DataFrame({"dept":["HR","IT","HR","IT"],"sal":[60,80,65,90]})
df["dept_avg"] = df.groupby("dept")["sal"].transform("mean")
print(df)

filter()

import pandas as pd

df = pd.DataFrame({"dept":["HR","IT","HR"],"sal":[60,80,65]})
# Keep only departments with mean salary > 70:
result = df.groupby("dept").filter(lambda g: g["sal"].mean() > 70)
print(result)

pivot_table

import pandas as pd

df = pd.DataFrame({
    "year":[2023,2023,2024,2024],
    "region":["East","West","East","West"],
    "sales":[100,150,120,200]
})
pt = pd.pivot_table(df, values="sales",
                    index="year", columns="region",
                    aggfunc="sum")
print(pt)

crosstab

import pandas as pd

df = pd.DataFrame({"gender":["M","F","M","F","M"],"result":["pass","pass","fail","fail","pass"]})
print(pd.crosstab(df["gender"], df["result"]))

apply()

import pandas as pd

df = pd.DataFrame({"dept":["HR","IT","HR"],"sal":[60,80,65]})

def range_sal(g):
    return g["sal"].max() - g["sal"].min()

print(df.groupby("dept").apply(range_sal))

استخدام unstack لإعادة التشكيل

import pandas as pd

df = pd.DataFrame({"year":[2023,2023,2024],"dept":["HR","IT","HR"],"sal":[60,80,65]})
result = df.groupby(["year","dept"])["sal"].mean().unstack()
print(result)

إعادة أخذ العينات للسلاسل الزمنية

import pandas as pd

idx = pd.date_range("2024-01-01", periods=90, freq="D")
df = pd.DataFrame({"sales": range(90)}, index=idx)
monthly = df.resample("ME").sum()
print(monthly)

تحقق سريع

مراجعة