الدمج والربط وتنظيف البيانات

pd.merge

import pandas as pd

users = pd.DataFrame({"id":[1,2,3],"name":["A","B","C"]})
orders = pd.DataFrame({"user_id":[1,1,2],"amount":[100,50,200]})
result = pd.merge(users, orders, left_on="id", right_on="user_id")
print(result)

أنواع الربط

import pandas as pd

A = pd.DataFrame({"key":["a","b","c"],"val":[1,2,3]})
B = pd.DataFrame({"key":["b","c","d"],"x":[10,20,30]})

print(pd.merge(A, B, on="key", how="left"))   # all of A
print(pd.merge(A, B, on="key", how="outer"))  # all rows

df.join

import pandas as pd

df1 = pd.DataFrame({"a":[1,2]}, index=["x","y"])
df2 = pd.DataFrame({"b":[3,4]}, index=["x","y"])
print(df1.join(df2))
# x: a=1 b=3
# y: a=2 b=4

concat

import pandas as pd

df1 = pd.DataFrame({"a":[1,2]})
df2 = pd.DataFrame({"a":[3,4]})
print(pd.concat([df1,df2], ignore_index=True))
# a: 1 2 3 4

# Horizontal:
df3 = pd.DataFrame({"b":[5,6]})
print(pd.concat([df1,df3], axis=1))

التعامل مع القيم المفقودة

import pandas as pd, numpy as np

df = pd.DataFrame({"a":[1,np.nan,3],"b":[np.nan,2,3]})
print(df.isna().sum())      # count nulls per column
df["a"] = df["a"].fillna(0)
df = df.dropna()            # drop rows with any null

استراتيجيات fillna

import pandas as pd, numpy as np

df = pd.DataFrame({"val":[1, np.nan, np.nan, 4]})
print(df["val"].ffill())   # forward fill: 1 1 1 4
print(df["val"].bfill())   # back fill:    1 4 4 4
print(df["val"].fillna(df["val"].mean()))   # fill with mean

تغيير أنواع البيانات

import pandas as pd

df = pd.DataFrame({"age":["25","30","35"],"date":["2024-01-01","2024-06-15","2024-12-31"]})
df["age"] = df["age"].astype(int)
df["date"] = pd.to_datetime(df["date"])
print(df.dtypes)

تنظيف السلاسل النصية

import pandas as pd

df = pd.DataFrame({"name":["  Alice  "," bob","CAROL"]})
df["clean"] = df["name"].str.strip().str.title()
print(df["clean"])   # Alice / Bob / Carol

إعادة تسمية الأعمدة وترتيبها

import pandas as pd

df = pd.DataFrame({"a":[1],"b":[2],"c":[3]})
df = df.rename(columns={"a":"alpha","b":"beta"})
df = df[["c","beta","alpha"]]   # reorder
print(df.columns.tolist())      # ['c', 'beta', 'alpha']

اكتشاف التكرارات

import pandas as pd

df = pd.DataFrame({"id":[1,2,1,3],"val":["a","b","a","c"]})
print(df.duplicated())           # [F F T F]
print(df[df.duplicated()])       # show duplicates
clean = df.drop_duplicates(subset=["id"])

استخدام apply لتحويل الصفوف والأعمدة

import pandas as pd

df = pd.DataFrame({"a":[1,2,3],"b":[4,5,6]})
# New column = row sum:
df["total"] = df.apply(lambda row: row["a"] + row["b"], axis=1)
print(df)

تحقق سريع

مراجعة