Python scatter plot and matplotlib

Question

I have a scatter plot in python onto which i want to draw a trendline. From the various examples i have found around the inte.net to learn how to draw the trendline my code is the following:

import matplotlib.pyplot as plt
import numpy as np

x=np.array([9.80,13.20,13.46,14.09,13.96,10.77,8.79,8.61,8.83,11.08,10.13,12.40,9.90,10.96,12.75,11.79,11.79,12.38,12.78,13.08,12.83,12.57,12.96,12.90,12.91,13.67,12.83,12.50,12.42,12.83,12.82,12.70,12.60,12.90,13.20])
y=np.array([0.0706,0.0969,0.0997,0.1031,0.0848,0.1044,0.0815,0.1030,0.0783,0.0970,0.1193,0.0796,0.0697,0.0738,0.0895,0.0912,0.0887,0.0973,0.0942,0.1052,0.0984,0.0965,0.0903,0.0876,0.1071,0.0872,0.0857,0.0967,0.0926,0.0837,0.0967,0.0935,0.0946,0.0930,0.0758
])

plt.scatter(x, y)
fit = np.polyfit(x, y, deg=4)
p = np.poly1d(fit)
plt.plot(x,p(x),"r--")
plt.show()

But the result line instead of being a curve line is just a mix of lines. Can anyone please explain my mistake?

Answer 1

I think it is because x values are not sorted. Look at this code:

import matplotlib.pyplot as plt 
import numpy as np

x=np.array([9.80,13.20,13.46,14.09,13.96,10.77,8.79,8.61,8.83,11.08,10.13,12.40,9.90,10.96,12.75,11.79,11.79,12.38,12.78,
            13.08,12.83,12.57,12.96,12.90,12.91,13.67,12.83,12.50,12.42,12.83,12.82,12.70,12.60,12.90,13.20]) 

y=np.array([0.0706,0.0969,0.0997,0.1031,0.0848,0.1044,0.0815,0.1030,0.0783,0.0970,0.1193,0.0796,0.0697,0.0738,
            0.0895,0.0912,0.0887,0.0973,0.0942,0.1052,0.0984,0.0965,0.0903,0.0876,0.1071,0.0872,0.0857,0.0967,0.0926,0.0837,0.0967,0.0935,0.0946,0.0930,0.0758 ])

# Here I sort x values and their corresponding y values
args = np.argsort(x)
x = x[args]
y = y[args]

plt.scatter(x, y) 
fit = np.polyfit(x, y, deg=4) 
p = np.poly1d(fit) 
plt.plot(x,p(x),"r--") 
plt.show()

Result:

Answer 2

This will do:

plt.scatter(x,y)
x = sorted(x) 
plt.plot(x,p(x),"r--") 

The trick is to sort x values before plotting line over them.

Answer 3

df["Medal"].fillna("None",inplace=True) df["Medal_Int"] = df["Medal"] df["Medal_Int"].replace(["Gold","Silver","Bronze","None"],[1,2,3,4],inplace=True) df_medals = df.groupby(["Year","Medal"]).count().reset_index()

df_entries = df.groupby(["Year"]).count().reset_index() moving_avg = df_entries.rolling(window=5).mean()

cum_sum[:5]

sns.scatterplot(x="Year",y="ID",size="Medal",hue="Medal",data=df_medals[df_medals["Medal"].="None"]) sns,lineplot(x="Year",y="ID",data=cum_sum)