I am trying to plot and linear fit the scatter (suppose 20) points. The plan is like this,
- take data from 20 csv files.
- scatter plot with label and legend(according to the file name).
- linear fit.
- perpendicular distance from the scatter point to the line.
I have done first two using this piece of code.
a_p=np.genfromtxt("a/stat.csv", delimiter=",", dtype=float)
a_r=np.genfromtxt("a/dis.csv", delimiter=",", dtype=float)
b_p=np.genfromtxt("b/stat.csv", delimiter=",", dtype=float)
b_r=np.genfromtxt("b/dis.csv", delimiter=" ", dtype=float)
plt.scatter(a_p[0]/a_p[8],(a_r[0]/a_r[2])**n, label="a")
plt.scatter(b_p[0]/b_p[8],(b_r[0]/b_r[2])**n, label="b")
#plt.plot(np.unique(x), np.poly1d(np.polyfit(x, y, 1))(np.unique(x)))
plt.legend()
Though there are many examples for linear fitting like this here, but they do declare two arrays x and y containing the points. But even if I store like those arrays and do the plot, I can't give customized legends.
EDIT:1 I have modified the code where, I have taken two arrays, x and y. Those arrays contain the points required for plot. However, getting wrong type of plot.
x=np.array([a_p[5]/a_p[2],b_p[5]/b_p[2],c_p[5]/c_p[2]],dtype=float)
y=np.array([(a_r[1]/a_r[2])**n,(b_r[1]/b_r[2])**n,(c_r[1]/c_r[2])**n],dtype=float)
plt.scatter(x,y)
m,b=np.polyfit(x,y,1)
plt.plot(x, y, x, np.array(x) * m+b)
plt.legend()
The plot is this,
Problem with the plot is, It is not scatter. Can I manually put legends to the scatter points by this method?
