Python+matplotlib实现折线图的美化

雪粒

1. 导入包

1. <br>import pandas as pd
2. import matplotlib.pyplot as plt
3. import matplotlib.ticker as ticker
4. import matplotlib.gridspec as gridspec

复制代码

1. [/code][size=5]2. 获得数据[/size]
3. [code]file_id = '1yM_F93NY4QkxjlKL3GzdcCQEnBiA2ltB'‘Python学习交流群：748989764 ’
4. url = f'https://drive.google.com/uc?id={file_id}'
5. df = pd.read_csv(url, index_col=0)
6. df

复制代码

数据长得是这样的：
 

 
3. 对数据做一些预处理

 按照需要，对数据再做一些预处理，代码及效果如下：

1. home_df = df.copy()
2. home_df = home_df.melt(id_vars = ["date", "home_team_name", "away_team_name"])
3. home_df["venue"] = "H"
4. home_df.rename(columns = {"home_team_name":"team", "away_team_name":"opponent"}, inplace = True)
5. home_df.replace({"variable":{"home_team_xG":"xG_for", "away_team_xG":"xG_ag"}}, inplace = True)

复制代码

1. away_df = df.copy()
2. away_df = away_df.melt(id_vars = ["date", "away_team_name", "home_team_name"])
3. away_df["venue"] = "A"
4. away_df.rename(columns = {"away_team_name":"team", "home_team_name":"opponent"}, inplace = True)
5. away_df.replace({"variable":{"away_team_xG":"xG_for", "home_team_xG":"xG_ag"}}, inplace = True)

复制代码

1. df = pd.concat([home_df, away_df]).reset_index(drop = True)
2. df

复制代码

 
 
 
4. 画图

1. # ---- Filter the data
3. Y_for = df[(df["team"] == "Lazio") & (df["variable"] == "xG_for")]["value"].reset_index(drop = True)
4. Y_ag = df[(df["team"] == "Lazio") & (df["variable"] == "xG_ag")]["value"].reset_index(drop = True)
5. X_ = pd.Series(range(len(Y_for)))
7. # ---- Compute rolling average
9. Y_for = Y_for.rolling(window = 5, min_periods = 0).mean() # min_periods is for partial avg.
10. Y_ag = Y_ag.rolling(window = 5, min_periods = 0).mean()

复制代码

1. fig, ax = plt.subplots(figsize = (7,3), dpi = 200)
3. ax.plot(X_, Y_for)
4. ax.plot(X_, Y_ag)

复制代码

 
 
 
使用matplotlib倒是可以快速把图画好了，但是太丑了。接下来进行优化。
 4.1 优化：添加点

 这里为每一个数据添加点
  

1. fig, ax = plt.subplots(figsize = (7,3), dpi = 200)
3. # --- Remove spines and add gridlines
5. ax.spines["left"].set_visible(False)
6. ax.spines["top"].set_visible(False)
7. ax.spines["right"].set_visible(False)
9. ax.grid(ls = "--", lw = 0.5, color = "#4E616C")
11. # --- The data
13. ax.plot(X_, Y_for, marker = "o")
14. ax.plot(X_, Y_ag, marker = "o")

复制代码

 

 
 
4.2 优化：设置刻度

 

1. fig, ax = plt.subplots(figsize = (7,3), dpi = 200)
3. # --- Remove spines and add gridlines
5. ax.spines["left"].set_visible(False)
6. ax.spines["top"].set_visible(False)
7. ax.spines["right"].set_visible(False)
9. ax.grid(ls = "--", lw = 0.25, color = "#4E616C")
11. # --- The data
13. ax.plot(X_, Y_for, marker = "o", mfc = "white", ms = 5)
14. ax.plot(X_, Y_ag, marker = "o", mfc = "white", ms = 5)
16. # --- Adjust tickers and spine to match the style of our grid
18. ax.xaxis.set_major_locator(ticker.MultipleLocator(2)) # ticker every 2 matchdays
19. xticks_ = ax.xaxis.set_ticklabels([x - 1 for x in range(0, len(X_) + 3, 2)])
20. # This last line outputs
21. # [-1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35]
22. # and we mark the tickers every two positions.
24. ax.xaxis.set_tick_params(length = 2, color = "#4E616C", labelcolor = "#4E616C", labelsize = 6)
25. ax.yaxis.set_tick_params(length = 2, color = "#4E616C", labelcolor = "#4E616C", labelsize = 6)
27. ax.spines["bottom"].set_edgecolor("#4E616C")

复制代码

 
 
4.3 优化：设置填充

 

1. fig, ax = plt.subplots(figsize = (7,3), dpi = 200)

复制代码

1. Python学习交流群：748989764

复制代码

1. <br>
2. # --- Remove spines and add gridlines
4. ax.spines["left"].set_visible(False)
5. ax.spines["top"].set_visible(False)
6. ax.spines["right"].set_visible(False)
8. ax.grid(ls = "--", lw = 0.25, color = "#4E616C")
10. # --- The data
12. ax.plot(X_, Y_for, marker = "o", mfc = "white", ms = 5)
13. ax.plot(X_, Y_ag, marker = "o", mfc = "white", ms = 5)
15. # --- Fill between
17. ax.fill_between(x = X_, y1 = Y_for, y2 = Y_ag, alpha = 0.5)
19. # --- Adjust tickers and spine to match the style of our grid
21. ax.xaxis.set_major_locator(ticker.MultipleLocator(2)) # ticker every 2 matchdays
22. xticks_ = ax.xaxis.set_ticklabels([x - 1 for x in range(0, len(X_) + 3, 2)])
24. ax.xaxis.set_tick_params(length = 2, color = "#4E616C", labelcolor = "#4E616C", labelsize = 6)
25. ax.yaxis.set_tick_params(length = 2, color = "#4E616C", labelcolor = "#4E616C", labelsize = 6)
27. ax.spines["bottom"].set_edgecolor("#4E616C")

复制代码

 
 
4.4 优化：设置填充颜色

1.当橙色线更高时，希望填充为橙色。但是上面的还无法满足，这里再优化一下.
  [code]fig, ax = plt.subplots(figsize = (7,3), dpi = 200)# --- Remove spines and add gridlinesax.spines["left"].set_visible(False)ax.spines["top"].set_visible(False)ax.spines["right"].set_visible(False)ax.grid(ls = "--", lw = 0.25, color = "#4E616C")# --- The dataax.plot(X_, Y_for, marker = "o", mfc = "white", ms = 5)ax.plot(X_, Y_ag, marker = "o", mfc = "white", ms = 5)# --- Fill between# Identify points where Y_for > Y_agpos_for = (Y_for > Y_ag)ax.fill_between(x = X_[pos_for], y1 = Y_for[pos_for], y2 = Y_ag[pos_for], alpha = 0.5)pos_ag = (Y_for