Pyplot: Plot a curve with ticks on one side

Question

Can I use the plot() function in matplotlib.pyplot to plot curves like this which have ticks on one side?:

Answer 1

Upgrade Based on the answer here I could extend the example:

def f(x): return x,  x * np.exp(-x*x)

def get_parameters(x,y):
    xp = 0.5*(x[1:nx] + x[0:nx-1])           # the points between
    yp = 0.5*(y[1:nx] + y[0:nx-1])
    dy = np.diff(y); dx = np.diff(x)         # the gradient
    nn = 40*np.sqrt(dx*dx + dy*dy)           # nn=norm; 40 = empirical hack for the normal shift
    dx = dx/nn;  dy = dy/nn                  # the components of the normals
    alpha = 180*np.arctan(dy/dx)/np.pi       # the slope angel to the normal
    return xp,yp,dx,dy,alpha

nx = 20;
ip = np.linspace(0,1,nx)
xr,yr = f(3*ip-0.5)                           # red front line
xb,yb = f(3*ip-0.5); yb = 0.7*yb -0.3         # blue front line

xpb, ypb, dx, dy, alphaB = get_parameters(xb,yb) # red points between
xpr, ypr,  _,  _, alphaR = get_parameters(xr,yr) # blue points between

plt.style.use('fast')  
fig, ax0 = plt.subplots(figsize=(20,20))
plt.plot(xr,yr, c='r', lw=5, label='warm front')
plt.plot(xb,yb, c='b', lw=5, label='cold front')

for j in range(nx-1):
    #--- set the blue markers ---
    marker_size_B = 900
    plt.scatter(xpb[j]-dy[j], ypb[j]+dx[j],
                s=marker_size_B, c='b', marker=(3, 0, alphaB[j]) )    

    #--- set the red markers ---
    marker_size_R=0.05
    halfR = mpl.patches.Wedge((xpr[j], ypr[j]), marker_size_R, theta1=0+alphaB[j], theta2=180+alphaB[j], color='r')
    ax0.add_artist(halfR)
plt.legend(prop={'size': 20})
ax0.set_aspect('equal'); plt.grid(); plt.margins(0.1);plt.show()

The densitiy of the markers can be controlled with marker_every.

Answer 2

Unfortunately I have not found a half circle marker for the cold front. But here a suggestion for the warm front:

def f(x): return x * np.exp(-x*x)

nx=20; x=np.linspace(-1,2.5,nx); y=f(x)  # the frontal line
xp = 0.5*(x[1:nx] + x[0:nx-1])           # the points between
yp = 0.5*(y[1:nx] + y[0:nx-1])
dy = np.diff(y); dx = np.diff(x)         # the gradient
nn = 40*np.sqrt(dx*dx + dy*dy)           # nn=norm; 40 = empirical hack for the normal shift
dx = dx/nn;  dy = dy/nn                  # the components of the normals
alpha = 180*np.arctan(dy/dx)/np.pi       # the slope angel to the normal

plt.style.use('fast')  
fig, ax0 = plt.subplots(figsize=(20,20))
xh = np.zeros_like(xp); yh = np.zeros_like(yp)
for j in range(nx-1):
    xh[j]=xp[j]-dy[j];  yh[j]=yp[j]+dx[j] # shift in the normal direction
    plt.scatter(xh[j],yh[j],s=900,c='r',marker=(3, 0, alpha[j]))

ax0.set_aspect('equal')                   # this is important !
plt.plot(x,y,c='r',lw=5, label='this is the frontal line')
plt.plot(xh,yh,ls='--', label='here are the markers')
plt.title("Important: set_aspect('equal')",fontsize=25, fontweight='bold')
plt.text(-1,0.2,'The markers have to be \n shifted in the normal direction \n of the frontal line')
plt.margins(0.1); plt.legend(prop={'size': 20});plt.grid(); plt.show()