我如何將 plot 多個圖形與 matplotlib 或 seaborn 合二為一

Question

我的dataframe是這樣的：點擊在線下載dataframe

在此處輸入 dataframe 的鏈接描述

我嘗試了以下代碼：

plt.plot(LessDF['DeptAvg'][LessDF['classes'] == 'COA111'], LessDF['week1'])
plt.plot(LessDF['DeptAvg'][LessDF['classes'] == 'COA111'], LessDF['week2'])
plt.plot(LessDF['DeptAvg'][LessDF['classes'] == 'COA111'], LessDF['week3'])

我得到了下面的 output，它只顯示了一行，以及我的代碼。

我想要 output 單獨的行，像這樣：

我怎樣才能用 matplotlib 或 seaborn 得到這個 output？

Answer 1

對於您應用的過濾器，DeptAvg 列中的所有值都是 67。

此外，您還提供了一個 boolean 作為您的 x: LessDF['DeptAvg'] == 'COA111' 。

此外，您在錯誤的列DeptAvg而不是classes上應用條件

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

df = pd.read_csv('../../../Desktop/LessDF.csv')
df_filtered = df[df['classes'] == 'COA111' ]

plt.plot(df_filtered['week1'],df_filtered['DeptAvg'],alpha=.5,)
plt.plot(df_filtered['week2'],df_filtered['DeptAvg'],alpha=.5)
plt.plot(df_filtered['week3'],df_filtered['DeptAvg'],alpha=.5)

plt.legend(['week1','week2','week3'])

plt.show()

更多信息在這里

Answer 2

對於 Seaborn，其object 接口從 v0.12 開始可用，您可以這樣做：

import pandas as pd
import seaborn as sns
import seaborn.objects as so

sns.set_theme()

首先，將數據框轉換為長格式，以便在第二張圖中進行處理。

df = pd.read_csv("LessDF.csv", index_col=0)

df_long = (
    # Convert to a long-form
    pd.melt(df,
        id_vars=["Id", "classes", "LessAvg", "DeptAvg"],
        var_name=["week"],
        value_name="point"
    )

    # Make `week1` to `1`
    .assign(week=lambda df_: df_.week.str.replace("week", ""))
)

然后

(
    so.Plot(
        # We don't have to drop rows but since `DeptAvg` doesn't change
        # over `classes` and `week`, we can de-duplicate them
        df_long.drop_duplicates(["classes", "week"]),
        x="week", y="DeptAvg", color="classes"
    )
    .add(so.Line())
    .limit(y=(0, 100))
)

如果您還想呈現每周的個人Id點，您可以這樣做：

(
    so.Plot(data=df_long, x="week", y="point", color="classes")
    .add(so.Dots(), so.Dodge(), so.Jitter(.3))
    .add(so.Line(linewidth=3, alpha=.8), y="DeptAvg")
    .limit(y=(0, 100))
)

Answer 3

# I done this using seaborn you can use matplotlib in between to code
plt.figure(figsize=(16, 16)) 
plt.subplot(no_of_rows, no_of_columns, plot_num)
plt.title('Any title 1')
sns.boxplot(df['column_name'])

Example :- we want 2 rows with columns of plots then we use
plt.subplot(2, 2, 1)
plt.title('Any title 1')
sns.distplot(df['column_name'], bins=20)

plt.subplot(2, 2, 2)
plt.title('Any title 2')
sns.distplot(df['column_name'], bins=20)

plt.subplot(2, 2, 3)
plt.title('Any title 3')
sns.distplot(df['column_name'], bins=20)

plt.subplot(2, 2, 4)
plt.title('Any title 4')
sns.distplot(df['column_name'], bins=20)

plt.show()

Answer 4

fig, ax = plt.subplots(2,3, figsize=(16,10))

sns.boxplot(df,y="arpu_6",ax=ax[0,0])
ax[0,0].set_title("June ARPU")
ax[0,0].set_ylim([0, 5000])

sns.boxplot(df,y="arpu_7",ax=ax[0,1])
ax[0,1].set_title("July ARPU")
ax[0,1].set_ylim([0, 5000])

sns.boxplot(df,y="arpu_8",ax=ax[0,2])
ax[0,2].set_title("Aug ARPU")
ax[0,2].set_ylim([0, 5000])

密碼是我自己筆記本上摘下來的，和你的dataframe不符，按你的需要修改即可。