英文:
Line plot for time series of grouped data in pandas
问题
我想要显示按客户类别分组的时间序列的折线图。以下是包含3个客户属于2个类别的4个连续月份的数据快照：
df = pd.DataFrame({'cust_id': [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3],'month': [200101, 200102, 200103, 200104, 200101, 200102, 200103, 200104, 200101, 200102, 200103, 200104],'volume': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12],'category': [1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2]})
如何绘制一张折线图（x轴上的时间段），显示这2组每个月的平均销售量？
英文:
I have monthly data for customers belonging to separate categories. I would like to display line plot of the time series grouped by customer category
Here is a snapshot containing data ("volume") over 4 consecutive months for 3 customers belonging to 2 categories
df = pd.DataFrame({'cust_id': [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3],'month': [200101, 200102, 200103, 200104, 200101, 200102, 200103, 200104, 200101, 200102, 200103, 200104],'volume': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12],'category': [1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2]})
How to plot a line plot (time period on the x axe) displaying the average monthly volume for each of these 2 groups?
答案1
得分: 2
关于使用 groupby.mean 和 seaborn.lineplot 的想法：
import seaborn as snsimport matplotlib.dates as mdates# 可选: 为了确保有一个分类调色板df['category'] = df['category'].astype('category')tmp = (df.groupby(['category', pd.to_datetime(df['month'], format='%Y%m')])['volume'].mean().reset_index(name='average volume'))ax = sns.lineplot(data=tmp, x='month', y='average volume', hue='category')# 更改标签ax.tick_params(axis='x', rotation=45)ax.xaxis.set_major_formatter(mdates.DateFormatter('%Y-%m'))
输出：
更改绘图宽度：
import matplotlib.pyplot as plt# ...fig, ax = plt.subplots(figsize=(15, 8))sns.lineplot(data=tmp, x='month', y='average volume', hue='category', ax=ax)# ...
输出：
英文:
What about using a groupby.mean and seaborn.lineplot:
import seaborn as snsimport matplotlib.dates as mdates# optional: to ensure having a categorical palettedf['category'] = df['category'].astype('category')tmp = (df.groupby(['category', pd.to_datetime(df['month'], format='%Y%m')])['volume'].mean().reset_index(name='average volume'))ax = sns.lineplot(data=tmp, x='month', y='average volume', hue='category')# change labelsax.tick_params(axis='x', rotation=45)ax.xaxis.set_major_formatter(mdates.DateFormatter('%Y-%m'))
Output:
Changing the plot width:
import matplotlib.pyplot as plt# ...fig, ax = plt.subplots(figsize=(15, 8))sns.lineplot(data=tmp, x='month', y='average volume', hue='category', ax=ax)# ...
Output:
答案2
得分: 1
pivot_table 提供了极少的代码就能控制所需的聚合形状。
你请求的时间序列：
import pandas as pdimport numpy as npTS = pd.pivot_table(data = df,values = ['volume'],index = ['month'],columns = ['category'],aggfunc = np.mean)
输出：
volumecategory 1 2month200101 1 7200102 2 8200103 3 9200104 4 10
确实，先将日期转换为datetime总是更好的选择，正如其他回答中所建议的。对于此操作，已经为mozway点赞：
df['month'] = pd.to_datetime(df['month'], format='%Y%m')
绘图如预期；可以添加任何你喜欢的装饰。
TS.plot(figsize=(24,8),ylabel = '月平均销量')
英文:
pivot_table offers great control in very few words over the desired aggregate shape.
Your requested time series:
import pandas as pdimport numpy as npTS = pd.pivot_table(data = df,values = ['volume'],index = ['month'],columns = ['category'],aggfunc = np.mean)
Output:
volumecategory 1 2month200101 1 7200102 2 8200103 3 9200104 4 10
Indeed, prior conversion to datetime always preferrable, as suggested by other respondants. Upvoted mozway for this:
df['month'] = pd.to_datetime(df['month'], format='%Y%m')
Plot is as expected; add any ornaments you like.
TS.plot(figsize=(24,8),ylabel = 'Monthly average volume')
答案3
得分: 0
此代码将创建一个折线图，其中 x 轴上是月份，y 轴上是体积，每条线对应一个客户类别。图例将显示与每条线相关的客户类别。
import pandas as pdimport matplotlib.pyplot as plt# Your DataFramedf = pd.DataFrame({'cust_id': [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3],'month': [200101, 200102, 200103, 200104, 200101, 200102, 200103, 200104, 200101, 200102, 200103, 200104],'volume': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12],'category': [1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2]})# Convert the 'month' column to a pandas datetime format for proper sortingdf['month'] = pd.to_datetime(df['month'], format='%y%m%d')# Group the data by 'category' and 'cust_id' and then pivot it to have 'category' as columnsgrouped_data = df.groupby(['category', 'cust_id', 'month'])['volume'].sum().unstack(level=0)# Plotting the dataplt.figure(figsize=(10, 6))for category in grouped_data.columns:plt.plot(grouped_data.index, grouped_data[category], label=f'Category {category}')plt.xlabel('Month')plt.ylabel('Volume')plt.title('Time Series of Volume Grouped by Customer Category')plt.legend()plt.show()
英文:
This code will create a line plot with the months on the x-axis, the volume on the y-axis, and each line corresponding to a customer category. The legend will show the customer category associated with each line.
<!-- begin snippet: js hide: false console: true babel: false -->
<!-- language: lang-python -->
import pandas as pdimport matplotlib.pyplot as plt# Your DataFramedf = pd.DataFrame({'cust_id': [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3],'month': [200101, 200102, 200103, 200104, 200101, 200102, 200103, 200104, 200101, 200102, 200103, 200104],'volume': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12],'category': [1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2]})# Convert the 'month' column to a pandas datetime format for proper sortingdf['month'] = pd.to_datetime(df['month'], format='%y%m%d')# Group the data by 'category' and 'cust_id' and then pivot it to have 'category' as columnsgrouped_data = df.groupby(['category', 'cust_id', 'month'])['volume'].sum().unstack(level=0)# Plotting the dataplt.figure(figsize=(10, 6))for category in grouped_data.columns:plt.plot(grouped_data.index, grouped_data[category], label=f'Category {category}')plt.xlabel('Month')plt.ylabel('Volume')plt.title('Time Series of Volume Grouped by Customer Category')plt.legend()plt.show()
<!-- end snippet -->
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论