diff --git a/homework/m1_numpy.py b/homework/m1_numpy.py
index 5ff41ab..8066b0e 100644
--- a/homework/m1_numpy.py
+++ b/homework/m1_numpy.py
@@ -19,19 +19,22 @@
 def green_mean():
     """建立 [10, 20, 30, 40, 50]，回傳所有元素的平均值 (float)"""
     # TODO: 你的程式碼
-    pass
+    arr = np.array([10, 20, 30, 40, 50])
+    return arr.mean()
 
 
 def green_double():
     """建立 [10, 20, 30, 40, 50]，回傳所有元素乘以 2 的 ndarray"""
     # TODO: 你的程式碼
-    pass
+    arr = np.array([10, 20, 30, 40, 50])
+    return arr*2
 
 
 def green_filter():
     """建立 [10, 20, 30, 40, 50]，回傳大於 25 的元素 (ndarray)"""
     # TODO: 你的程式碼
-    pass
+    arr = np.array([10, 20, 30, 40, 50])
+    return arr[arr > 25]
 
 
 # ============================================================
@@ -42,7 +45,8 @@ def green_filter():
 def yellow_expensive_count(prices):
     """回傳單價 > 1000 的商品數量 (int)"""
     # TODO: 你的程式碼
-    pass
+    prices_arr = np.array(prices)
+    return len(prices_arr[prices_arr > 1000])
 
 
 def yellow_top3_stock_indices(stocks):
@@ -51,7 +55,9 @@ def yellow_top3_stock_indices(stocks):
     提示：np.argsort
     """
     # TODO: 你的程式碼
-    pass
+    stocks_arr = np.array(stocks)
+    stocks_arr = np.argsort(stocks)[::-1]
+    return stocks_arr[:3]
 
 
 def yellow_restock_cost(prices, stocks):
@@ -60,7 +66,8 @@ def yellow_restock_cost(prices, stocks):
     提示：布林遮罩 + .sum()
     """
     # TODO: 你的程式碼
-    pass
+    total = prices[prices < 500] *50
+    return total.sum()
 
 
 # ============================================================
@@ -77,4 +84,8 @@ def red_double11_prices(prices, stocks):
     提示：np.where 可以巢狀使用
     """
     # TODO: 你的程式碼
-    pass
+    prices_07 = prices[stocks >= 100] * 0.7
+    prices_09 = prices[(stocks >= 20) & (stocks < 100)] * 0.9
+    prices_original = prices[stocks < 20]
+    final_price = np.concatenate([prices_original, prices_07, prices_09])
+    return final_price
diff --git a/homework/m2_pandas_cleaning.py b/homework/m2_pandas_cleaning.py
index fa36bff..cc6ee97 100644
--- a/homework/m2_pandas_cleaning.py
+++ b/homework/m2_pandas_cleaning.py
@@ -20,7 +20,9 @@ def green_read_csv():
     提示：pd.read_csv()
     """
     # TODO: 你的程式碼
-    pass
+    DATA = "/datasets/ecommerce/orders_raw.csv"
+    df = pd.read_csv(DATA)
+    return df
 
 
 def green_shape(df):
@@ -29,7 +31,8 @@ def green_shape(df):
     提示：df.shape
     """
     # TODO: 你的程式碼
-    pass
+    
+    return df.shape
 
 
 def green_dtypes(df):
@@ -38,7 +41,8 @@ def green_dtypes(df):
     提示：df.dtypes
     """
     # TODO: 你的程式碼
-    pass
+
+    return df.dtypes
 
 
 # ============================================================
@@ -52,7 +56,9 @@ def yellow_clean_columns(df):
     提示：df.columns.str.strip().str.lower()
     """
     # TODO: 你的程式碼
-    pass
+
+    df.columns = df.columns.str.strip().str.lower()
+    return df
 
 
 def yellow_clean_amount(df):
@@ -63,7 +69,12 @@ def yellow_clean_amount(df):
     提示：.str.replace() + .astype(float)
     """
     # TODO: 你的程式碼
-    pass
+    arr = (df['amount']
+                .astype(str)
+                .str.replace('$', '', regex=False)
+                .str.replace(',', '', regex=False)
+                .astype(float))
+    return arr
 
 
 def yellow_drop_duplicates(df):
@@ -72,7 +83,9 @@ def yellow_drop_duplicates(df):
     提示：df.drop_duplicates()
     """
     # TODO: 你的程式碼
-    pass
+
+    arr = df.drop_duplicates()
+    return arr
 
 
 # ============================================================
@@ -93,4 +106,16 @@ def red_clean_orders(path):
     提示：pd.to_datetime(errors='coerce')
     """
     # TODO: 你的程式碼
-    pass
+
+    df = pd.read_csv(path)
+    df.columns = df.columns.str.strip().str.lower()
+    df["amount"] = (df['amount']
+                .astype(str)
+                .str.replace('$', '', regex=False)
+                .str.replace(',', '', regex=False)
+                .astype(float))
+    df["order_date"] = pd.to_datetime(df['order_date'], errors='coerce')
+    df = df.dropna(subset=['order_date'])
+    df['qty'] = df['qty'].fillna(df['qty'].median())
+    df = df.drop_duplicates()
+    return df
\ No newline at end of file
diff --git a/homework/m3_pandas_advanced.py b/homework/m3_pandas_advanced.py
index 68410e8..11fbed1 100644
--- a/homework/m3_pandas_advanced.py
+++ b/homework/m3_pandas_advanced.py
@@ -24,19 +24,42 @@ def green_load_and_merge():
     提示：pd.merge(how='left')
     """
     # TODO: 你的程式碼
-    pass
-
+    orders = pd.read_csv("datasets/ecommerce/orders_clean.csv")
+    customers = pd.read_csv("datasets/ecommerce/customers.csv")
+    products = pd.read_csv("datasets/ecommerce/products.csv")
+    df = (
+            orders
+            .merge(customers, on = 'customer_id', how='left')
+            .merge(products, on = 'product_id', how='left')
+    )
+    return df
 
 def green_row_count(df):
     """回傳 DataFrame 的列數 (int)"""
     # TODO: 你的程式碼
-    pass
+    orders = pd.read_csv("datasets/ecommerce/orders_clean.csv")
+    customers = pd.read_csv("datasets/ecommerce/customers.csv")
+    products = pd.read_csv("datasets/ecommerce/products.csv")
+    df = (
+            orders
+            .merge(customers, on = 'customer_id', how='left')
+            .merge(products, on = 'product_id', how='left')
+    )
+    return df.shape[0]
 
 
 def green_column_list(df):
     """回傳 DataFrame 的所有欄位名稱 (list)"""
     # TODO: 你的程式碼
-    pass
+    orders = pd.read_csv("datasets/ecommerce/orders_clean.csv")
+    customers = pd.read_csv("datasets/ecommerce/customers.csv")
+    products = pd.read_csv("datasets/ecommerce/products.csv")
+    df = (
+            orders
+            .merge(customers, on = 'customer_id', how='left')
+            .merge(products, on = 'product_id', how='left')
+    )
+    return df.columns
 
 
 # ============================================================
@@ -50,7 +73,9 @@ def yellow_top_category(df):
     提示：groupby('category')['amount'].sum()
     """
     # TODO: 你的程式碼
-    pass
+
+    amount = df.groupby('category')['amount'].sum().sort_values(ascending=False).head(1)
+    return amount
 
 
 def yellow_gold_vip_stats(df):
@@ -60,7 +85,12 @@ def yellow_gold_vip_stats(df):
     提示：df[df['vip_level'] == 'Gold']
     """
     # TODO: 你的程式碼
-    pass
+
+    vip = df[df['vip_level'] == 'Gold']
+    vip_sum = len(vip['order_id'])
+    vip_amount = float(vip['amount'].sum())
+    Gold_Vip = (vip_sum, vip_amount)
+    return Gold_Vip
 
 
 def yellow_region_avg_amount(df):
@@ -70,7 +100,9 @@ def yellow_region_avg_amount(df):
     提示：groupby('region')['amount'].mean()
     """
     # TODO: 你的程式碼
-    pass
+
+    Series = df.groupby('region')['amount'].mean()
+    return Series
 
 
 # ============================================================
@@ -94,4 +126,38 @@ def red_rfm_top5(df):
     提示：groupby('customer_id').agg(...)
     """
     # TODO: 你的程式碼
-    pass
+    orders = pd.read_csv("datasets/ecommerce/orders_clean.csv")
+    customers = pd.read_csv("datasets/ecommerce/customers.csv")
+    products = pd.read_csv("datasets/ecommerce/products.csv")
+    # R = orders.groupby('customer_id')['order_date'].max()
+    # F = orders.groupby('customer_id')['order_id'].count()
+    # M = orders.groupby('customer_id')['amount'].sum().sort_values(ascending=False).head()
+    # df = (
+    #             orders
+    #             .merge(customers, on = 'customer_id', how='left')
+    #             .merge(products, on = 'product_id', how='left')
+    #     )
+    RFM = (
+    orders.groupby('customer_id')
+    .agg(
+        recency = ('order_date', 'max'),
+        frequency = ('order_id', 'count'),
+        monetary = ('amount', 'sum'),
+    )
+    .reset_index()
+)
+    rfm_named = RFM.merge(
+    customers[['customer_id', 'customer_name']],
+    on = 'customer_id',
+    how='left',
+)
+
+    final_5 = (
+    rfm_named
+    .sort_values("M", ascending=False)
+    .head(5)
+    .reset_index(drop=True)
+    [['customer_id', 'customer_name', 'R', 'F', 'M']]
+)
+
+    return final_5
\ No newline at end of file
diff --git a/homework/m4_timeseries.py b/homework/m4_timeseries.py
index 047af6a..9306551 100644
--- a/homework/m4_timeseries.py
+++ b/homework/m4_timeseries.py
@@ -27,7 +27,11 @@ def green_avg_by_month():
     提示：df['order_date'].dt.month
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+
+    result = df.groupby(df["order_date"].dt.month)['amount'].mean().round(1)
+
+    return result
 
 
 def green_top3_dates():
@@ -37,7 +41,9 @@ def green_top3_dates():
     提示：value_counts().head(3)
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    top3 = df['order_date'].dt.date.value_counts().head(3)
+    return top3
 
 
 def green_date_range():
@@ -46,7 +52,11 @@ def green_date_range():
     格式為 pandas Timestamp
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    time_1 = df['order_date'].min()
+    time_2 = df['order_date'].max()
+    result = (time_1, time_2)
+    return result
 
 
 # ============================================================
@@ -60,7 +70,9 @@ def yellow_monthly_revenue():
     提示：set_index('order_date').resample('ME')['amount'].sum()
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    test = df.set_index('order_date').resample('ME')['amount'].sum()
+    return test
 
 
 def yellow_rolling_avg(monthly_revenue):
@@ -71,7 +83,8 @@ def yellow_rolling_avg(monthly_revenue):
     提示：.rolling(window=3).mean()
     """
     # TODO: 你的程式碼
-    pass
+
+    return monthly_revenue.rolling(window=3).mean()
 
 
 def yellow_category_median(df):
@@ -81,7 +94,8 @@ def yellow_category_median(df):
     提示：groupby + median + sort_values
     """
     # TODO: 你的程式碼
-    pass
+
+    return df.groupby('category')['amount'].median().sort_values(ascending=False)
 
 
 # ============================================================
@@ -101,4 +115,18 @@ def red_monthly_report():
     提示：resample + agg + pct_change
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    df['year_mon'] = df['order_date'].dt.to_period('M')
+    monthly_report = (df.groupby('year_mon')
+                      .agg(
+                        當月訂單數 = ('order_id', 'count'),
+                        當月總營收 = ('amount', 'sum'),
+                        單月不重複客戶數 = ('customer_id', 'nunique'),
+                      )
+                      .sort_index()
+                      )
+    monthly_report['客單價'] = (monthly_report['當月總營收'] / monthly_report['當月訂單數'])
+    monthly_report['月營收成長率'] = (monthly_report['當月總營收'].pct_change()*100)
+    monthly_report = monthly_report.reset_index().rename(columns={'year_mon' : '月份'})
+    monthly_report['月份'] = monthly_report['月份'].astype(str)
+    return monthly_report
diff --git a/homework/m5_visualization.py b/homework/m5_visualization.py
index 7e7335d..38b4cee 100644
--- a/homework/m5_visualization.py
+++ b/homework/m5_visualization.py
@@ -29,7 +29,19 @@ def green_bar_category():
     提示：sns.countplot 或 value_counts().plot.bar()
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    counts = df['category'].value_counts().reset_index()
+    counts.columns = ['category', 'order_count']
+    figure = plt.figure(figsize=(8, 4))
+    sns.barplot(
+    data=counts,
+    x='category', y='order_count',
+    hue='category', palette='viridis', legend=False,
+    )
+
+    plt.tight_layout()
+    return figure
+
 
 
 def green_hist_amount():
@@ -39,7 +51,14 @@ def green_hist_amount():
     提示：sns.histplot(bins=20) 或 plt.hist()
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    figure = plt.figure(figsize=(9, 4))
+    sns.histplot(data=df, x='amount', bins=20, kde=True, color='green')
+    plt.title('Order Amount Distribution', fontweight='bold')
+    plt.xlabel('Amount (NT$)')
+    plt.ylabel('Frequency')
+    plt.tight_layout()
+    return figure
 
 
 def green_set_labels():
@@ -51,7 +70,23 @@ def green_set_labels():
     回傳 matplotlib Figure 物件
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    counts = df['category'].value_counts().reset_index()
+    counts.columns = ['category', 'order_count']
+    figure = plt.figure(figsize=(8, 4))
+    sns.barplot(
+    data=counts,
+    x='category', y='order_count',
+    hue='category', palette='viridis', legend=False,
+    )
+    plt.title('Order Count by Category', fontweight='bold')
+    plt.xlabel('Category')
+    plt.ylabel('Order Count')
+    for i, v in enumerate(counts['order_count']):
+        plt.text(i, v, f'{v:,}', ha='center', va='bottom', fontsize=10)
+
+    plt.tight_layout()
+    return figure
 
 
 # ============================================================
@@ -68,7 +103,27 @@ def yellow_line_region_trend():
     提示：分別 groupby 再 plot，或用 sns.lineplot(hue='region')
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    ns_df = df[df['region'].isin(['North', 'South'])]
+    monthly_ns = (
+    ns_df.groupby(['month', 'region'])['amount']
+    .sum()
+    .reset_index()
+    )
+
+    figure = plt.figure(figsize=(10, 4))
+    sns.lineplot(
+    data=monthly_ns,
+    x='month', y='amount',
+    hue='region', marker='o', linewidth=2,
+)
+    plt.title('Monthly Revenue: North vs South', fontweight='bold')
+    plt.xlabel('Month')
+    plt.ylabel('Revenue (NT$)')
+    plt.xticks(rotation=45)
+    plt.legend(title='Region')
+    plt.tight_layout()
+    return figure
 
 
 def yellow_box_vip():
@@ -78,7 +133,18 @@ def yellow_box_vip():
     提示：sns.boxplot(x='vip_level', y='amount', data=df)
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    figure = plt.figure(figsize=(9, 5))
+    sns.boxplot(
+    data=df,
+    x='vip_level', y='amount',
+    hue='vip_level', palette='Set3', legend=False,
+    )
+    plt.title('Order Amount Distribution by VIP Level', fontweight='bold')
+    plt.xlabel('VIP Level')
+    plt.ylabel('Amount (NT$)')
+    plt.tight_layout()
+    return figure
 
 
 def yellow_scatter_price_amount():
@@ -88,7 +154,17 @@ def yellow_scatter_price_amount():
     提示：plt.scatter() 或 sns.scatterplot()
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    figure = plt.figure(figsize=(9, 5))
+    sns.scatterplot(
+    data=df,
+    x='unit_price', y='amount',
+    )
+    plt.title('Unit_price Amount Distribution by Category', fontweight='bold')
+    plt.xlabel('Unit Price')
+    plt.ylabel('Amount (NT$)')
+    plt.tight_layout()
+    return figure
 
 
 # ============================================================
@@ -107,4 +183,55 @@ def red_category_dashboard(category="Electronics"):
     提示：fig, axes = plt.subplots(2, 2, figsize=(14, 10))
     """
     # TODO: 你的程式碼
-    pass
+    df = _load_data()
+    elec = df[df['category'] == category].copy()
+    elec_monthly = elec.groupby('month')['amount'].sum().reset_index()
+    elec_region = (
+    elec.groupby('region')['amount']
+    .sum()
+    .sort_values(ascending=False)
+    .reset_index()
+    )
+    elec_top5 = (
+    elec.groupby('product_name')['amount']
+    .sum()
+    .sort_values(ascending=False)
+    .head(5)
+    .reset_index()
+    )
+    fig, axes = plt.subplots(2, 2, figsize=(14, 10))
+    fig.suptitle('Electronics Category Dashboard', fontsize=16, fontweight='bold')
+    sns.lineplot(
+    data=elec_monthly, x='month', y='amount',
+    marker='o', linewidth=2, color='#1f77b4', ax=axes[0, 0],
+    )
+    axes[0, 0].set_title('Monthly Revenue Trend')
+    axes[0, 0].set_xlabel('Month')
+    axes[0, 0].set_ylabel('Revenue (NT$)')
+    axes[0, 0].tick_params(axis='x', rotation=45)
+    sns.barplot(
+    data=elec_region, x='region', y='amount',
+    hue='region', palette='viridis', legend=False, ax=axes[0, 1],
+)
+    axes[0, 1].set_title('Revenue by Region')
+    axes[0, 1].set_xlabel('Region')
+    axes[0, 1].set_ylabel('Revenue (NT$)')
+    for i, v in enumerate(elec_region['amount']):
+        axes[0, 1].text(i, v, f'{v:,.0f}', ha='center', va='bottom', fontsize=9)
+    sns.barplot(
+    data=elec_top5, y='product_name', x='amount',
+    hue='product_name', palette='magma', legend=False, ax=axes[1, 0],
+    )
+    axes[1, 0].set_title('Top 5 Products')
+    axes[1, 0].set_xlabel('Revenue (NT$)')
+    axes[1, 0].set_ylabel('Product')
+    sns.histplot(
+    data=elec, x='amount', bins=25, kde=True,
+    color='#d62728', ax=axes[1, 1],
+    )
+    axes[1, 1].set_title('Amount Distribution')
+    axes[1, 1].set_xlabel('Amount (NT$)')
+    axes[1, 1].set_ylabel('Frequency')
+
+    plt.tight_layout()
+    return fig
diff --git a/homework/m6_plotly_capstone.py b/homework/m6_plotly_capstone.py
index 0e2c32a..4b8329c 100644
--- a/homework/m6_plotly_capstone.py
+++ b/homework/m6_plotly_capstone.py
@@ -26,7 +26,15 @@ def green_plotly_bar():
     提示：px.bar()
     """
     # TODO: 你的程式碼
-    pass
+    df = pd.read_csv(
+    'datasets/ecommerce/orders_enriched.csv',
+    parse_dates=['order_date'],)
+    region_rev = df.groupby('category', as_index=False)['amount'].sum().sort_values('amount', ascending=False)
+    fig = px.bar(region_rev, x='category', y='amount', text='amount',
+             color='category', title='Revenue by Category')
+    fig.update_traces(texttemplate='%{text:,.0f}', textposition='outside')
+    fig.update_layout(height=400, showlegend=False)
+    return fig
 
 
 def green_plotly_line():
@@ -37,7 +45,16 @@ def green_plotly_line():
     提示：先 groupby 月份算總營收，再 px.line()
     """
     # TODO: 你的程式碼
-    pass
+    df = pd.read_csv(
+    'datasets/ecommerce/orders_enriched.csv',
+    parse_dates=['order_date'],)
+    df['month'] = df['order_date'].dt.to_period('M').astype(str)
+    monthly = df.groupby('month', as_index=False)['amount'].sum()
+
+    fig = px.line(monthly, x='month', y='amount', markers=True,
+              title='Monthly Revenue Trend')
+    fig.update_layout(height=400)
+    return fig
 
 
 def green_plotly_pie():
@@ -48,7 +65,14 @@ def green_plotly_pie():
     提示：px.pie()
     """
     # TODO: 你的程式碼
-    pass
+    df = pd.read_csv(
+    'datasets/ecommerce/orders_enriched.csv',
+    parse_dates=['order_date'],)
+    vip_rev = df.groupby('vip_level', as_index=False)['amount'].sum()
+    fig = px.pie(vip_rev, names='vip_level', values='amount',
+             title='VIP Level Share', hole=0.4)
+    fig.update_layout(height=400)
+    return fig
 
 
 # ============================================================
@@ -63,7 +87,29 @@ def yellow_clean_and_merge(raw_path, customers_path, products_path):
     回傳：合併後的 DataFrame
     """
     # TODO: 你的程式碼
-    pass
+    df = pd.read_csv(
+    raw_path,
+    parse_dates=['order_date'],)
+    df.columns = df.columns.str.strip().str.lower()
+    df['amount'] = (
+        df['amount'].astype(str)
+        .str.replace('$', '', regex=False)
+        .str.replace(',', '', regex=False)
+        .astype(float)
+    )
+    df['order_date'] = pd.to_datetime(df['order_date'], errors='coerce')
+    df = df.dropna(subset=['order_date'])
+    df['qty'] = df['qty'].fillna(df['qty'].median())
+    df = df.drop_duplicates()
+    customers = pd.read_csv(customers_path)
+    products  = pd.read_csv(products_path)
+
+    enriched = (
+    df
+    .merge(customers, on='customer_id', how='left')
+    .merge(products,  on='product_id',  how='left')
+)
+    return enriched
 
 
 def yellow_kpi_summary(df):
@@ -77,7 +123,35 @@ def yellow_kpi_summary(df):
     }
     """
     # TODO: 你的程式碼
-    pass
+    df = pd.read_csv('datasets/ecommerce/orders_raw.csv',
+    parse_dates=['order_date'],)
+    customers = pd.read_csv('datasets/ecommerce/customers.csv')
+    products  = pd.read_csv('datasets/ecommerce/products.csv')
+    df.columns = df.columns.str.strip().str.lower()
+    df['amount'] = (
+        df['amount'].astype(str)
+        .str.replace('$', '', regex=False)
+        .str.replace(',', '', regex=False)
+        .astype(float)
+    )
+    df['order_date'] = pd.to_datetime(df['order_date'], errors='coerce')
+    df = df.dropna(subset=['order_date'])
+    df['qty'] = df['qty'].fillna(df['qty'].median())
+    df = df.drop_duplicates()
+    
+    enriched = (
+    df
+    .merge(customers, on='customer_id', how='left')
+    .merge(products,  on='product_id',  how='left')
+    )
+    kpis = {
+    '總營收':     enriched['amount'].sum(),
+    '總訂單數':   len(enriched),
+    '活躍顧客數': enriched['customer_id'].nunique(),
+    '客單價':     enriched['amount'].sum() / len(enriched),
+    }
+
+    return kpis
 
 
 def yellow_plotly_scatter(df):
@@ -91,7 +165,15 @@ def yellow_plotly_scatter(df):
     提示：px.scatter(hover_data=['product_name'])
     """
     # TODO: 你的程式碼
-    pass
+    fig = px.scatter(
+        df,
+        x='unit_price',
+        y='amount',
+        color='category',
+        hover_data=['product_name'],
+        title='Unit Price vs. Amount by Category'
+    )
+    return fig
 
 
 # ============================================================
@@ -115,4 +197,54 @@ def red_dashboard():
     提示：from plotly.subplots import make_subplots
     """
     # TODO: 你的程式碼
-    pass
+    df = pd.read_csv('datasets/ecommerce/orders_raw.csv',
+    parse_dates=['order_date'],)
+    customers = pd.read_csv('datasets/ecommerce/customers.csv')
+    products  = pd.read_csv('datasets/ecommerce/products.csv')
+    df.columns = df.columns.str.strip().str.lower()
+    df['amount'] = (
+        df['amount'].astype(str)
+        .str.replace('$', '', regex=False)
+        .str.replace(',', '', regex=False)
+        .astype(float)
+    )
+    df['order_date'] = pd.to_datetime(df['order_date'], errors='coerce')
+    df = df.dropna(subset=['order_date'])
+    df['qty'] = df['qty'].fillna(df['qty'].median())
+    df = df.drop_duplicates()
+    
+    enriched = (
+    df
+    .merge(customers, on='customer_id', how='left')
+    .merge(products,  on='product_id',  how='left')
+    )
+    monthly     = enriched.groupby('month', as_index=False)['amount'].sum()
+    top_prod    = (enriched.groupby('product_name', as_index=False)['amount']
+               .sum().sort_values('amount', ascending=False).head(10))
+    region_rev  = enriched.groupby('region', as_index=False)['amount'].sum()
+    cat_rev     = enriched.groupby('category', as_index=False)['amount'].sum()
+    fig = make_subplots(
+    rows=2, cols=2,
+    subplot_titles=('Monthly Revenue Trend',
+                    'Top 10 Products',
+                    'Revenue by Region',
+                    'Category Share'),
+    specs=[[{'type': 'xy'},     {'type': 'xy'}],
+           [{'type': 'xy'},     {'type': 'domain'}]],
+    )
+
+    fig.add_trace(go.Scatter(x=monthly['month'], y=monthly['amount'],
+                         mode='lines+markers', name='Monthly'), row=1, col=1)
+    fig.add_trace(go.Bar(x=top_prod['product_name'], y=top_prod['amount'],
+                     name='Top Products'), row=1, col=2)
+    fig.add_trace(go.Bar(x=region_rev['region'], y=region_rev['amount'],
+                     name='Region'), row=2, col=1)
+    fig.add_trace(go.Pie(labels=cat_rev['category'], values=cat_rev['amount'],
+                     name='Category', hole=0.4), row=2, col=2)
+
+    fig.update_layout(
+    title_text='<b>E-Commerce Sales Dashboard — 2025【解答版】</b>',
+    height=750, showlegend=False,
+    )
+    fig.update_xaxes(tickangle=45, row=1, col=2)
+    return fig