😃[3週-Day 3]😃

データをMatplotlibに可視化

データをより美しく表現しましょう.

1.Matplotlibの起動

2.よく使うPlottingオプション

サイズ:figsize

タイトル:title

ラベル:_label

目盛り:_tics

例:legend

3. Matplotlib Case Study

折れ線図(Plot)

散点図

箱図(Box Plot)

棒グラフ

円形図(Pie Chart)

4.Theかっこいい図、Seaborn Case Study

カーネル密度図

計数図

猫図(Cat Plot)

棒グラフ(Strip Plot)

熱図(Heatmap)

I.Matplotlibの起動

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

%matplotlib inline

plt.plot([2,4,2,4,2])
plt.show()

# plotting 할 도면을 선언합니다.
plt.figure(figsize=(6,6))  # 사이즈를 조절해줄 수 있습니다.
plt.plot([0,1,2,3,4])
plt.show()


# 2차 함수 그래프
x = np.array([1, 2, 3,  4,  5]) # 정의역
y = np.array([1, 4, 9, 16, 25]) # f(x)
plt.plot(x, y)
plt.show()

x = np.arange(-10, 10, 0.01)
plt.plot(x, x**2)
plt.show()

x = np.arange(-10, 10, 0.01)
plt.xlabel("x value")
plt.ylabel("f(x) value")

plt.plot(x, x**2)
plt.show()![](https://media.vlpt.us/images/donchanee/post/756d200d-66a0-4a22-b99e-8069c474b297/output_5_0.png)

x = np.arange(-10, 10, 0.01)
plt.xlabel("x value")
plt.ylabel("f(x) value")

plt.axis([-5, 5, 0, 25])

plt.plot(x, x**2)
plt.show()

# x, y 축에 눈금 설정하기

x = np.arange(-10, 10, 0.01)
plt.xlabel("x value")
plt.ylabel("f(x) value")
plt.axis([-5, 5, 0, 25])

plt.xticks([i for i in range(-5, 6)])
plt.yticks([i for i in range(0, 25, 3)])

plt.title("y = x^2 graph")

plt.plot(x, x**2, label="trend")
plt.legend()
plt.show()

II. Matplotlib Case Study

-折れ線グラフ(Plot)

#꺾은선 그래프 (Plot) - 시계열 데이터에 활용

x = np.arange(20)
y = np.random.randint(0,20,20)

plt.plot(x, y)

# Extra: y축을 20까지, 5단위로 표시하기
plt.axis([0,20,0,20])
plt.yticks([0,5,10,15,20])

plt.show()

-散点図

# 산점도 (Scatter plot) - x와 y의 상관관계를 파악하는데에 활용

plt.scatter(x,y)
plt.show()

-ブロック図(Box Plot)

# 박스 그림 (Box Plot) - 수치형 데이터에 대한 정보

plt.boxplot(y)
plt.show()

plt.boxplot((x,y))
plt.title("Box plot of x, y")
plt.show()

-棒グラフ(Bar Chart)

# 막대 그래프 (Bar plot)

plt.bar(x,y)

# Extra: xtics를 올바르게 처리해봅시다.

plt.xticks(np.arange(0,20,1))
plt.show()

# Histogram - 도수분표를 직사각형의 막대 형태지만, 계급이 존재합니다.
# 0, 1, 2가 아니라 0~2 까지의 "범주형" 데이터로 구성

plt.hist(y, bins=np.arange(0,20,2))
plt.xticks(np.arange(0,20,2))
plt.show()

-円形図(Pie Chart)

# 원형 그래프 (Pie chart)

z = [100, 300, 200, 400]
plt.pie(z, labels=['A', 'B', 'C', 'D'])
plt.show()

III.The素晴らしいグラフ、Seaborn Case Studio

import seaborn as sns

# histogram

x = np.arange(0,22,2)
y = np.random.randint(0,20,20)

plt.hist(y, bins=x)
plt.show()

-カーネル密度図

#kdeplot

sns.kdeplot(y, shade=True)
plt.show()

-カウントチャート（Count Plot）

# 카운트 플롯(Count Plot)

vote_df = pd.DataFrame({"name":['A', 'B', 'C'], "vote":[True, True, False]})

vote_df

name
vote
0
A
True
1
B
True
2
C
False

vote_count = vote_df.groupby('vote').count()

vote_count

plt.bar(x=[False, True], height=vote_count['name'])
plt.show()

sns.countplot(x = vote_df['vote'])

plt.show()

-ケイト画像(Cat Plot)

# 캣 플롯 (Cat Plot)

covid = pd.read_csv("./country_wise_latest.csv")

s = sns.catplot(x='WHO Region', y='Confirmed', data=covid)
s.fig.set_size_inches(10,6)
plt.show()

-棒グラフ(Strip Plot)

# 스트립 플롯 (Strip plot)

sns.stripplot(x='WHO Region', y='Recovered', data=covid)
plt.show()

# 스웜 플롯 (Swarm plot)

sns.swarmplot(x='WHO Region', y='Recovered', data=covid)
plt.show()

/usr/local/lib/python3.7/site-packages/seaborn/categorical.py:1296: UserWarning: 22.7% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.
  warnings.warn(msg, UserWarning)
/usr/local/lib/python3.7/site-packages/seaborn/categorical.py:1296: UserWarning: 69.6% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.
  warnings.warn(msg, UserWarning)
/usr/local/lib/python3.7/site-packages/seaborn/categorical.py:1296: UserWarning: 79.2% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.
  warnings.warn(msg, UserWarning)
/usr/local/lib/python3.7/site-packages/seaborn/categorical.py:1296: UserWarning: 54.3% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.
  warnings.warn(msg, UserWarning)
/usr/local/lib/python3.7/site-packages/seaborn/categorical.py:1296: UserWarning: 31.2% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.
  warnings.warn(msg, UserWarning)

-ホットマップ(Heatmap)

# 히트맵 (Heatmap)

sns.heatmap(covid.corr())
plt.show()