Adding line to scatter plot using python's matplotlib

This draws a diagonal line which is independent of the scatter plot data and which stays rooted to the axes even if you resize the window:

import numpy as np
import matplotlib.pyplot as plt
import matplotlib.lines as mlines
import matplotlib.transforms as mtransforms

x, y = np.random.random((2, 100))*2
fig, ax = plt.subplots()
ax.scatter(x, y, c='black')
line = mlines.Line2D([0, 1], [0, 1], color='red')
transform = ax.transAxes
line.set_transform(transform)
ax.add_line(line)
plt.show()

Besides unutbu's answer one other option is to get the limits of the axis after you ploted the data and to use them to add the line. After this you will still need to change back the axis limits as they would change with the addition of the line:

# Scatter Plot
x = data_calc_hourly.temp
y =  data_obs_hourly.temp

lineStart = data_calc_hourly.temp.min() 
lineEnd = data_calc_hourly.temp.max()  

plt.figure()
plt.scatter(x, y, color = 'k', alpha=0.5)
y_lim = plt.ylim()
x_lim = plt.xlim()
plt.plot(x_lim, y_lim, 'k-', color = 'r')
plt.ylim(y_lim)
plt.xlim(x_lim)
plt.show()