Trực Quan Hóa Dữ Liệu

1import matplotlib.pyplot as plt
2import numpy as np
3import pandas as pd
4
5# Style đẹp hơn mặc định
6plt.style.use('seaborn-v0_8-whitegrid')

1# Cách 1: Quick plot
2plt.figure(figsize=(10, 6))
3plt.plot([1, 2, 3, 4], [10, 20, 25, 30])
4plt.title("Quick Plot")
5plt.show()
6
7# Cách 2: Object-oriented (KHUYẾN KHÍCH)
8fig, ax = plt.subplots(figsize=(10, 6))
9ax.plot([1, 2, 3, 4], [10, 20, 25, 30])
10ax.set_title("OOP Style")
11ax.set_xlabel("X")
12ax.set_ylabel("Y")
13plt.show()
14
15# Cách 3: Multiple subplots
16fig, axes = plt.subplots(2, 2, figsize=(12, 10))
17axes[0, 0].plot([1,2,3], [1,4,9])
18axes[0, 1].bar(['A','B','C'], [3,7,5])
19axes[1, 0].scatter(np.random.rand(50), np.random.rand(50))
20axes[1, 1].hist(np.random.randn(1000), bins=30)
21plt.tight_layout()
22plt.show()

1fig, axes = plt.subplots(2, 3, figsize=(15, 10))
2
3# Line Plot
4x = np.linspace(0, 10, 100)
5axes[0,0].plot(x, np.sin(x), label='sin(x)')
6axes[0,0].plot(x, np.cos(x), label='cos(x)', linestyle='--')
7axes[0,0].legend()
8axes[0,0].set_title("Line Plot")
9
10# Bar Chart
11axes[0,1].bar(['Q1','Q2','Q3','Q4'], [100, 150, 120, 180], color='steelblue')
12axes[0,1].set_title("Bar Chart")
13
14# Scatter Plot
15axes[0,2].scatter(np.random.rand(50), np.random.rand(50), 
16                   c=np.random.rand(50), s=np.random.rand(50)*500, alpha=0.6)
17axes[0,2].set_title("Scatter Plot")
18
19# Histogram
20axes[1,0].hist(np.random.randn(1000), bins=30, edgecolor='black')
21axes[1,0].set_title("Histogram")
22
23# Pie Chart
24axes[1,1].pie([30, 25, 20, 15, 10], labels=['A','B','C','D','E'], autopct='%1.1f%%')
25axes[1,1].set_title("Pie Chart")
26
27# Box Plot
28data = [np.random.normal(0, std, 100) for std in range(1, 4)]
29axes[1,2].boxplot(data, labels=['σ=1', 'σ=2', 'σ=3'])
30axes[1,2].set_title("Box Plot")
31
32plt.suptitle("Matplotlib Basics", fontsize=16, fontweight='bold')
33plt.tight_layout()
34plt.show()

1import seaborn as sns
2
3sns.set_theme(style="whitegrid")
4
5# Load sample dataset
6tips = sns.load_dataset("tips")
7print(tips.head())

1fig, axes = plt.subplots(2, 2, figsize=(12, 10))
2
3# Histogram + KDE
4sns.histplot(data=tips, x="total_bill", kde=True, ax=axes[0,0])
5axes[0,0].set_title("Histogram + KDE")
6
7# KDE Plot grouped
8sns.kdeplot(data=tips, x="total_bill", hue="time", fill=True, ax=axes[0,1])
9axes[0,1].set_title("KDE by Time")
10
11# Box Plot
12sns.boxplot(data=tips, x="day", y="total_bill", hue="time", ax=axes[1,0])
13axes[1,0].set_title("Box Plot")
14
15# Violin Plot
16sns.violinplot(data=tips, x="day", y="total_bill", hue="sex", 
17               split=True, ax=axes[1,1])
18axes[1,1].set_title("Violin Plot")
19
20plt.tight_layout()
21plt.show()

1fig, axes = plt.subplots(1, 3, figsize=(15, 5))
2
3# Bar Plot (mean + CI)
4sns.barplot(data=tips, x="day", y="total_bill", hue="sex", 
5            errorbar="sd", ax=axes[0])
6axes[0].set_title("Bar Plot (Mean ± SD)")
7
8# Count Plot
9sns.countplot(data=tips, x="day", hue="time", palette="Set2", ax=axes[1])
10axes[1].set_title("Count Plot")
11
12# Swarm Plot
13sns.swarmplot(data=tips, x="day", y="total_bill", hue="sex", ax=axes[2])
14axes[2].set_title("Swarm Plot")
15
16plt.tight_layout()
17plt.show()

1# Scatter với nhiều dimensions
2plt.figure(figsize=(10, 6))
3sns.scatterplot(data=tips, x="total_bill", y="tip",
4                hue="time", size="size", style="sex",
5                palette="viridis", sizes=(20, 200))
6plt.title("Total Bill vs Tip (Multi-dimensional)")
7plt.show()
8
9# Regression plot
10sns.lmplot(data=tips, x="total_bill", y="tip", 
11           hue="smoker", col="time", height=5)
12plt.show()

1plt.figure(figsize=(8, 6))
2numeric_cols = tips.select_dtypes(include=[np.number])
3corr = numeric_cols.corr()
4
5sns.heatmap(corr, annot=True, cmap="coolwarm", center=0,
6            fmt=".2f", linewidths=0.5, square=True)
7plt.title("Correlation Heatmap")
8plt.show()

1# Tất cả cặp biến
2sns.pairplot(tips, hue="time", diag_kind="kde", corner=True)
3plt.show()

1import plotly.express as px
2
3# Sample data
4df = px.data.gapminder()
5tips = px.data.tips()

1fig = px.scatter(df.query("year == 2007"),
2                 x="gdpPercap", y="lifeExp",
3                 color="continent", size="pop",
4                 hover_name="country",
5                 log_x=True,
6                 title="GDP vs Life Expectancy (2007)")
7fig.show()

1# Grouped bar
2fig = px.bar(tips, x="day", y="total_bill", color="sex",
3             barmode="group",
4             title="Total Bill by Day and Sex")
5fig.show()

1fig = px.histogram(tips, x="total_bill", nbins=30,
2                   color="time",
3                   marginal="box",  # hoặc "rug", "violin"
4                   title="Distribution of Total Bill")
5fig.show()

1fig = px.box(tips, x="day", y="total_bill", color="smoker",
2             notched=True,
3             title="Total Bill Distribution")
4fig.show()

1fig = px.choropleth(df.query("year == 2007"),
2                    locations="iso_alpha",
3                    color="lifeExp",
4                    hover_name="country",
5                    color_continuous_scale="Viridis",
6                    title="Life Expectancy (2007)")
7fig.show()

1fig = px.scatter(df, x="gdpPercap", y="lifeExp",
2                 animation_frame="year",
3                 animation_group="country",
4                 size="pop", color="continent",
5                 hover_name="country",
6                 log_x=True, size_max=55,
7                 range_x=[100, 100000],
8                 range_y=[25, 90],
9                 title="Gapminder Animation")
10fig.show()

1import plotly.graph_objects as go
2from plotly.subplots import make_subplots
3
4fig = go.Figure()
5
6fig.add_trace(go.Scatter(
7    x=[1, 2, 3, 4], y=[10, 11, 12, 13],
8    mode='lines+markers', name='Line 1'
9))
10fig.add_trace(go.Scatter(
11    x=[1, 2, 3, 4], y=[12, 9, 15, 12],
12    mode='lines+markers', name='Line 2'
13))
14
15fig.update_layout(
16    title="Custom Plot",
17    xaxis_title="X", yaxis_title="Y"
18)
19fig.show()

1fig = go.Figure()
2
3fig.add_trace(go.Scatter(
4    x=[1,2,3,4], y=[100,200,300,400],
5    name="Revenue ($)", yaxis="y1"
6))
7fig.add_trace(go.Bar(
8    x=[1,2,3,4], y=[10,20,15,25],
9    name="Orders", yaxis="y2", opacity=0.5
10))
11
12fig.update_layout(
13    yaxis=dict(title="Revenue ($)", side="left"),
14    yaxis2=dict(title="Orders", side="right", overlaying="y"),
15    title="Dual Y-Axis"
16)
17fig.show()

1fig = make_subplots(rows=2, cols=2,
2                    subplot_titles=("Scatter", "Bar", "Line", "Histogram"))
3
4fig.add_trace(go.Scatter(x=[1,2,3], y=[4,5,6], mode='markers'), row=1, col=1)
5fig.add_trace(go.Bar(x=['A','B','C'], y=[3,7,5]), row=1, col=2)
6fig.add_trace(go.Scatter(x=[1,2,3], y=[1,2,3], mode='lines'), row=2, col=1)
7fig.add_trace(go.Histogram(x=np.random.randn(500)), row=2, col=2)
8
9fig.update_layout(height=600, title_text="Subplots")
10fig.show()

1# Interactive HTML
2fig.write_html("chart.html")
3
4# Static image
5fig.write_image("chart.png", scale=2)  # Cần kaleido
6fig.write_image("chart.pdf")

1# Themes: darkgrid, whitegrid, dark, white, ticks
2sns.set_theme(style="whitegrid", palette="Set2",
3              font_scale=1.2)

1# Categorical: Set1, Set2, Dark2, Paired, husl
2# Sequential: Blues, Reds, Viridis, coolwarm
3# Custom
4custom = ["#FF6B6B", "#4ECDC4", "#45B7D1", "#96CEB4"]
5sns.set_palette(custom)

1# plotly, plotly_white, plotly_dark, ggplot2, seaborn, simple_white
2fig = px.scatter(tips, x="total_bill", y="tip", 
3                 template="plotly_white")

1# Matplotlib
2fig, ax = plt.subplots()
3ax.plot(x, y)
4plt.show()
5
6# Seaborn
7sns.histplot(data=df, x="col", hue="cat", kde=True)
8sns.scatterplot(data=df, x="x", y="y", hue="cat")
9sns.heatmap(corr, annot=True)
10sns.pairplot(df)
11
12# Plotly
13fig = px.scatter(df, x="x", y="y", color="cat", size="val")
14fig = px.bar(df, x="cat", y="val", color="group")
15fig.show()