【Python】pandasの使い方完全ガイド｜基礎からデータ分析まで最速マスター

# pandasのインストール
pip install pandas

# すでにインストール済みの場合、最新版にアップグレード
pip install -U pandas

# バージョン確認
python -c "import pandas as pd; print(pd.__version__)"

import pandas as pd

# バージョンを表示して、インポートできているか確認
print(pd.__version__)

import pandas as pd

# リストからSeriesを作成
s = pd.Series([10, 20, 30], index=["A", "B", "C"])

print("Series本体:")
print(s)

print("\nvalues(値の部分):", s.values)
print("index(インデックス):", s.index)

print(s["A"])    # ラベルで参照
print(s[0])      # 位置で参照

import pandas as pd

# 辞書からDataFrameを作成
data = {
    "name": ["Alice", "Bob", "Charlie"],
    "age": [25, 30, 35],
    "city": ["Tokyo", "Osaka", "Nagoya"],
}

df = pd.DataFrame(data)

print(df)
print("\n列名:", df.columns)
print("インデックス:", df.index)

import pandas as pd

# CSVファイルの読み込み
df = pd.read_csv("sales.csv")  # 同じフォルダにsales.csvがある想定

# 先頭5行を表示
print(df.head())

# 代表的なオプション指定例
df = pd.read_csv(
    "sales.csv",
    encoding="utf-8",  # 文字コード(cp932などに変更することもある)
    sep=",",           # 区切り文字(タブ区切りなら"\t")
    header=0,          # 何行目をヘッダ行とみなすか
)

import pandas as pd

# Excelファイルから読み込み
df = pd.read_excel("sales.xlsx", sheet_name="Sheet1")

print(df.head())

# DataFrameをExcelに出力
df.to_excel("sales_output.xlsx", sheet_name="result", index=False)

import pandas as pd
import sqlite3

# JSONの読み込み
df_json = pd.read_json("data.json")

# SQLデータベースからの読み込み
conn = sqlite3.connect("mydb.sqlite")

# SQLクエリでデータ取得
df_sql = pd.read_sql_query("SELECT * FROM sales", conn)

# DataFrameをSQLテーブルとして保存
df_sql.to_sql("sales_backup", conn, if_exists="replace", index=False)
conn.close()

import pandas as pd

df = pd.DataFrame(
    {
        "name": ["Alice", "Bob", "Charlie"],
        "age": [25, 30, 35],
        "city": ["Tokyo", "Osaka", "Nagoya"],
    }
)

# 列の選択
age_col = df["age"]            # Seriesとして取得
subset_cols = df[["name", "age"]]  # 複数列の選択はリストで

print(age_col)
print("\n---")
print(subset_cols)

# loc: ラベルベースのアクセス
print(df.loc[0])             # インデックスラベル0の行
print(df.loc[0:1, ["name"]]) # 行0〜1の"name"列

# iloc: 位置ベースのアクセス
print(df.iloc[0])            # 0番目の行
print(df.iloc[0:2, 0:2])     # 0〜1行、0〜1列

import pandas as pd

df = pd.DataFrame(
    {
        "price": [100, 200, 300],
        "quantity": [1, 3, 2],
    }
)

# 新しい列 total を追加
df["total"] = df["price"] * df["quantity"]

print(df)

# 列の削除(axis=1 または columns引数)
df_dropped = df.drop(columns=["quantity"])

print(df_dropped)

df_renamed = df.rename(columns={"price": "unit_price", "quantity": "qty"})
print(df_renamed)

import pandas as pd

df = pd.DataFrame(
    {
        "id": [101, 102, 103],
        "name": ["Alice", "Bob", "Charlie"],
        "score": [80, 90, 85],
    }
)

# name列をインデックスとして設定
df_indexed = df.set_index("name")

print(df_indexed)

df_reset = df_indexed.reset_index()
print(df_reset)

import pandas as pd
import numpy as np

df = pd.DataFrame(
    {
        "name": ["Alice", "Bob", "Charlie"],
        "age": [25, np.nan, 35],
        "score": [80, 90, np.nan],
    }
)

print("元のDataFrame:")
print(df)

# 欠損の有無を確認
print("\n欠損値の有無:")
print(df.isna())

# 行ごと削除(dropna)
print("\n欠損値を含む行を削除:")
print(df.dropna())

# 平均値で補完(fillna)
age_mean = df["age"].mean()
score_mean = df["score"].mean()

df_filled = df.fillna({"age": age_mean, "score": score_mean})

print("\n平均値で補完したDataFrame:")
print(df_filled)

import pandas as pd

df = pd.DataFrame(
    {
        "date": ["2024-01-01", "2024-01-02", "2024-01-03"],
        "amount": ["1000", "2000", "1500"],  # 文字列として読み込まれた金額
    }
)

print("元のdtypes:")
print(df.dtypes)

# 日付型に変換
df["date"] = pd.to_datetime(df["date"])

# 数値型に変換
df["amount"] = df["amount"].astype(int)

print("\n変換後のdtypes:")
print(df.dtypes)

import pandas as pd

df = pd.DataFrame(
    {
        "id": [1, 2, 2, 3],
        "name": ["Alice", "Bob", "Bob", "Charlie"],
    }
)

print("元のDataFrame:")
print(df)

# 重複の有無を確認(最初の出現をFalse、重複分をTrueとする)
print("\nduplicated():")
print(df.duplicated())

# 完全に同じ行を削除
df_unique = df.drop_duplicates()

print("\n重複行を削除:")
print(df_unique)

# 特定の列だけを基準に重複判定
df_unique_name = df.drop_duplicates(subset=["name"])
print("\nname列を基準に重複削除:")
print(df_unique_name)

import pandas as pd

df = pd.DataFrame(
    {
        "customer": ["Alice", "Bob", "Alice", "Bob", "Charlie"],
        "amount": [1000, 2000, 1500, 3000, 4000],
    }
)

print("元のデータ:")
print(df)

# customerごとにamountの合計を計算
grouped = df.groupby("customer")["amount"].sum()

print("\n顧客別の合計金額:")
print(grouped)

import pandas as pd

df = pd.DataFrame(
    {
        "customer": ["Alice", "Bob", "Alice", "Bob", "Charlie"],
        "amount": [1000, 2000, 1500, 3000, 4000],
        "items": [1, 2, 1, 3, 4],
    }
)

# 顧客ごとに、合計・平均・件数を計算
agg_df = df.groupby("customer").agg(
    total_amount=("amount", "sum"),
    mean_amount=("amount", "mean"),
    order_count=("amount", "count"),
    total_items=("items", "sum"),
)

print(agg_df)

import pandas as pd

df = pd.DataFrame(
    {
        "date": ["2024-01-01", "2024-01-01", "2024-01-02", "2024-01-02"],
        "product": ["A", "B", "A", "B"],
        "region": ["East", "East", "West", "West"],
        "amount": [1000, 2000, 1500, 3000],
    }
)

# 商品×地域ごとの売上合計
pivot = pd.pivot_table(
    df,
    index="product",
    columns="region",
    values="amount",
    aggfunc="sum",
)

print(pivot)

import pandas as pd

df1 = pd.DataFrame({"id": [1, 2], "value": ["A", "B"]})
df2 = pd.DataFrame({"id": [3, 4], "value": ["C", "D"]})

# 縦結合(axis=0)
df_vert = pd.concat([df1, df2], axis=0, ignore_index=True)

print("縦結合:")
print(df_vert)

df_left = pd.DataFrame({"id": [1, 2], "value1": ["A", "B"]})
df_right = pd.DataFrame({"id": [1, 2], "value2": ["X", "Y"]})

# インデックスを揃えた上で横結合
df_horiz = pd.concat([df_left.set_index("id"), df_right.set_index("id")], axis=1)

print("横結合:")
print(df_horiz)

import pandas as pd

customers = pd.DataFrame(
    {
        "customer_id": [1, 2, 3],
        "name": ["Alice", "Bob", "Charlie"],
    }
)

orders = pd.DataFrame(
    {
        "order_id": [101, 102, 103],
        "customer_id": [1, 2, 2],
        "amount": [1000, 2000, 1500],
    }
)

# customer_idをキーとして結合
merged = pd.merge(customers, orders, on="customer_id", how="inner")

print(merged)

# 左外部結合の例
merged_left = pd.merge(customers, orders, on="customer_id", how="left")
print(merged_left)

import pandas as pd

df1 = pd.DataFrame({"value1": [10, 20, 30]}, index=["A", "B", "C"])
df2 = pd.DataFrame({"value2": [100, 200, 300]}, index=["A", "B", "D"])

joined = df1.join(df2, how="left")  # 左外部結合

print(joined)

import pandas as pd

df = pd.DataFrame(
    {
        "name": ["Charlie", "Alice", "Bob"],
        "age": [35, 25, 30],
    }
)

# ageで昇順にソート
sorted_age = df.sort_values(by="age")

# nameで降順にソート
sorted_name_desc = df.sort_values(by="name", ascending=False)

print("age昇順:")
print(sorted_age)
print("\nname降順:")
print(sorted_name_desc)

df_indexed = df.set_index("name")
print(df_indexed.sort_index())

import pandas as pd

df = pd.DataFrame(
    {
        "name": ["Alice", "Bob", "Charlie"],
        "age": [25, 30, 35],
        "city": ["Tokyo", "Osaka", "Tokyo"],
    }
)

# ageが30以上
cond1 = df["age"] >= 30
result1 = df[cond1]

# ageが30以上 かつ cityがTokyo
cond2 = (df["age"] >= 30) & (df["city"] == "Tokyo")
result2 = df[cond2]

print("age >= 30:")
print(result1)
print("\nage >= 30 & city == 'Tokyo':")
print(result2)

import pandas as pd

df = pd.DataFrame(
    {
        "name": ["Alice", "Bob", "Charlie"],
        "age": [25, 30, 35],
        "city": ["Tokyo", "Osaka", "Tokyo"],
    }
)

# queryで条件抽出
result = df.query("age >= 30 and city == 'Tokyo'")

print(result)

import pandas as pd

# 日次データを作成
df = pd.DataFrame(
    {
        "date": pd.date_range("2024-01-01", periods=10, freq="D"),
        "sales": [100, 120, 80, 90, 110, 130, 150, 170, 160, 140],
    }
)

# 日付をインデックスに設定
df = df.set_index("date")

print("日次データ:")
print(df.head())

# 月次にリサンプリングして合計を計算
monthly = df.resample("M").sum()

print("\n月次合計:")
print(monthly)

import pandas as pd

df = pd.DataFrame(
    {
        "date": pd.date_range("2024-01-01", periods=7, freq="D"),
        "sales": [100, 120, 80, 90, 110, 130, 150],
    }
).set_index("date")

# 3日移動平均
df["sales_ma3"] = df["sales"].rolling(window=3).mean()

print(df)

import pandas as pd

df = pd.DataFrame(
    {
        "date": pd.date_range("2024-01-01", periods=5, freq="D"),
        "sales": [100, 120, 90, 110, 130],
    }
).set_index("date")

# 1日前の売上を追加
df["sales_prev"] = df["sales"].shift(1)

# 前日比(差分)
df["diff"] = df["sales"] - df["sales_prev"]

print(df)

import pandas as pd
import matplotlib.pyplot as plt

df = pd.DataFrame(
    {
        "date": pd.date_range("2024-01-01", periods=5, freq="D"),
        "sales": [100, 120, 90, 110, 130],
    }
).set_index("date")

# Seriesのplot
df["sales"].plot(title="Daily Sales")
plt.xlabel("Date")
plt.ylabel("Sales")
plt.tight_layout()
plt.show()

import pandas as pd
import matplotlib.pyplot as plt

df = pd.DataFrame(
    {
        "category": ["A", "B", "C"],
        "value": [10, 20, 15],
    }
)

# 棒グラフ
df.plot(kind="bar", x="category", y="value", title="Bar Chart")
plt.tight_layout()
plt.show()

import pandas as pd
import matplotlib.pyplot as plt

s = pd.Series([10, 12, 10, 15, 13, 12, 11, 14])

# ヒストグラム
s.plot(kind="hist", bins=5, title="Histogram")
plt.xlabel("Value")
plt.tight_layout()
plt.show()

import pandas as pd
import matplotlib.pyplot as plt

df = pd.DataFrame(
    {
        "date": pd.date_range("2024-01-01", periods=5, freq="D"),
        "sales": [100, 120, 90, 110, 130],
        "cost": [60, 70, 50, 65, 80],
    }
).set_index("date")

ax = df[["sales", "cost"]].plot(
    kind="line",
    title="Sales & Cost Over Time",
    figsize=(6, 4),
    marker="o",
)

ax.set_xlabel("Date")
ax.set_ylabel("Amount")
ax.legend(loc="upper left")
plt.grid(True)
plt.tight_layout()
plt.show()

import pandas as pd

# 1. データ読み込み
df = pd.DataFrame(
    {
        "order_id": [1, 2, 3, 4, 5, 6],
        "date": [
            "2024-01-01",
            "2024-01-01",
            "2024-01-02",
            "2024-01-02",
            "2024-01-03",
            "2024-01-03",
        ],
        "customer": ["Alice", "Bob", "Alice", "Charlie", "Bob", "Alice"],
        "product": ["A", "A", "B", "A", "B", "B"],
        "amount": [1000, 2000, 1500, 3000, 2500, 1800],
    }
)

# 2. 型変換(日付)
df["date"] = pd.to_datetime(df["date"])

# 3-1. 顧客別売上合計
customer_sales = df.groupby("customer")["amount"].sum().sort_values(ascending=False)

# 3-2. 商品×日付のピボット(売上合計)
product_daily = pd.pivot_table(
    df,
    index="date",
    columns="product",
    values="amount",
    aggfunc="sum",
)

print("顧客別売上合計:")
print(customer_sales)
print("\n商品×日付の売上:")
print(product_daily)

import pandas as pd

# サンプルのアクセスログデータ
df = pd.DataFrame(
    {
        "ip": [
            "10.0.0.1",
            "10.0.0.2",
            "10.0.0.1",
            "10.0.0.3",
            "10.0.0.2",
            "10.0.0.1",
        ],
        "datetime": [
            "2024-01-01 09:15",
            "2024-01-01 09:20",
            "2024-01-01 10:05",
            "2024-01-01 10:10",
            "2024-01-01 11:00",
            "2024-01-01 11:30",
        ],
        "url": ["/", "/products", "/", "/contact", "/products", "/"],
    }
)

# 日時をdatetime型に変換し、インデックスに設定
df["datetime"] = pd.to_datetime(df["datetime"])
df = df.set_index("datetime")

# 1. 時間帯別アクセス数(1時間ごと)
hourly = df["ip"].resample("H").count()

# 2. URL別アクセス数ランキング
url_counts = df["url"].value_counts()

print("時間帯別アクセス数:")
print(hourly)

print("\nURL別アクセス数:")
print(url_counts)