def plot_evaluate_active_subspace_density_1d_step(
line, points_for_eval_in_interval, rotated_vertices, density_fn,
points_for_eval, cnt_all, W, mapped_vertices, density_vals):
f, axs = plt.subplots(1, 2, sharey=False, figsize=(16, 6))
axs[0].plot(line[0, :], line[1, :], '^b-', ms=10)
axs[0].plot(points_for_eval_in_interval, 0, 'rs')
axs[0].plot(rotated_vertices[0, :], rotated_vertices[1, :], 'o-k')
axs[0].plot(rotated_vertices[0, :], rotated_vertices[1, :], 'o-k')
ss_samples = np.vstack(
(points_for_eval[:cnt_all], np.zeros(cnt_all, float)))
axs[0].plot(ss_samples[0, :], ss_samples[1, :], 'r-')
I = [0, 3]
axs[0].plot(rotated_vertices[0, I], rotated_vertices[1, I], 'o-k')
rotated_density_fn = lambda x: density_fn(np.dot(W.T, x))
limits = [
rotated_vertices[0, :].min(), rotated_vertices[0, :].max(),
rotated_vertices[1, :].min(), rotated_vertices[1, :].max()
]
X, Y, Z = get_meshgrid_function_data(rotated_density_fn, limits, 51)
xx = np.vstack((X.flatten()[np.newaxis, :], Y.flatten()[np.newaxis, :]))
I = np.where(np.absolute(np.dot(W.T, xx)) > 1)[1]
Z = Z.flatten()
levels = np.linspace(Z.min(), Z.max(), 30)
Z[I] = np.nan
cmap = plt.cm.coolwarm
cmap.set_bad('white', 1.)
Z = Z.reshape(X.shape[0], X.shape[1])
#axs[0].imshow(Z[::-1,:],extent=limits,cmap=cmap)
axs[0].contourf(X, Y, Z, extent=limits, cmap=cmap, levels=levels)
axs[1].set_xlim([mapped_vertices.min(), mapped_vertices.max()])
axs[1].set_ylim([0, 2])
axs[1].plot(points_for_eval[:cnt_all], density_vals[:cnt_all], 'k')
def plot_tensor_product_lagrange_basis_2d(level, ii, jj, ax=None):
abscissa, tmp = clenshaw_curtis_pts_wts_1D(level)
abscissa_1d = [abscissa, abscissa]
barycentric_weights_1d = [compute_barycentric_weights_1d(abscissa_1d[0]),
compute_barycentric_weights_1d(abscissa_1d[1])]
training_samples = cartesian_product(abscissa_1d, 1)
fn_vals = np.zeros((training_samples.shape[1], 1))
idx = jj*abscissa_1d[1].shape[0]+ii
fn_vals[idx] = 1.
def f(samples): return multivariate_barycentric_lagrange_interpolation(
samples, abscissa_1d, barycentric_weights_1d, fn_vals,
np.array([0, 1]))
plot_limits = [-1, 1, -1, 1]
num_pts_1d = 101
X, Y, Z = get_meshgrid_function_data(f, plot_limits, num_pts_1d)
if ax is None:
ax = create_3d_axis()
cmap = mpl.cm.coolwarm
plot_surface(X, Y, Z, ax, axis_labels=None, limit_state=None,
alpha=0.3, cmap=mpl.cm.coolwarm, zorder=3, plot_axes=False)
num_contour_levels = 30
offset = -(Z.max()-Z.min())/2
cmap = mpl.cm.gray
ax.contourf(
X, Y, Z, zdir='z', offset=offset,
levels=np.linspace(Z.min(), Z.max(), num_contour_levels),
cmap=cmap, zorder=-1)
ax.plot(training_samples[0, :], training_samples[1, :],
offset*np.ones(training_samples.shape[1]), 'o',
zorder=100, color='b')
x = np.linspace(-1, 1, 100)
y = training_samples[1, idx]*np.ones((x.shape[0]))
z = f(np.vstack((x[np.newaxis, :], y[np.newaxis, :])))[:, 0]
ax.plot(x, Y.max()*np.ones((x.shape[0])), z, '-r')
ax.plot(abscissa_1d[0], Y.max()*np.ones(
(abscissa_1d[0].shape[0])), np.zeros(abscissa_1d[0].shape[0]), 'or')
y = np.linspace(-1, 1, 100)
x = training_samples[0, idx]*np.ones((y.shape[0]))
z = f(np.vstack((x[np.newaxis, :], y[np.newaxis, :])))[:, 0]
ax.plot(X.min()*np.ones((x.shape[0])), y, z, '-r')
ax.plot(X.min()*np.ones(
(abscissa_1d[1].shape[0])), abscissa_1d[1],
np.zeros(abscissa_1d[1].shape[0]), 'or')
def plot_2d(function, num_XX_test_1d, bounds, XX_train_values=None, ax=None):
#gp_mean_func = lambda XX: gp_predict(gp,x_kernel,kernel_ff,XX.T)[0]
gp_mean_func = lambda XX: function(XX.T)
num_contour_levels = 20
if ax is None:
fig, ax = plt.subplots(1, 1, figsize=(8, 6))
X, Y, Z = get_meshgrid_function_data(gp_mean_func, bounds, num_XX_test_1d)
cset = ax.contourf(X,
Y,
Z,
levels=np.linspace(Z.min(), Z.max(),
num_contour_levels))
if XX_train_values is not None:
ax.plot(XX_train_values[:, 0], XX_train_values[:, 1], 'ko')
plt.colorbar(cset, ax=ax)
return ax
def plot(fun, ax):
X, Y, Z = get_meshgrid_function_data(fun, plot_limits, num_samples_1d)
dx = (plot_limits[1] - plot_limits[0]) / num_samples_1d
dy = (plot_limits[3] - plot_limits[2]) / num_samples_1d
print(('integral of func', Z.sum() *
(dx * dy))) # should converge to 1 as num_samples_1d->\infty
z_max = np.percentile(Z.flatten(), 99.9)
if Z.max() / z_max < 10:
z_max = Z.max()
levels = np.linspace(0, z_max, num_contour_levels)
cset = ax.contourf(X, Y, Z, levels=levels, cmap=mpl.cm.coolwarm)
ax.set_xlim(0, 1)
ax.set_ylim(0, 1)
ax.set_xlabel('$z_1$')
ax.set_ylabel('$z_2$')
def plot_optimization_objective_and_constraints_2D(constraints, objective,
plot_limits):
from pyapprox.visualization import get_meshgrid_function_data
num_pts_1d = 100
num_contour_levels = 30
fig, axs = plt.subplots(1, 3, figsize=(3 * 8, 6))
#for ii in range(len(constraint_functions)+1):
for ii in range(len(constraints.constraints) + 1):
#if ii==len(constraint_functions):
if ii == len(constraints.constraints):
function = objective
else:
# def function(design_samples):
# vals = np.empty((design_samples.shape[1]))
# for jj in range(design_samples.shape[1]):
# vals[jj]=constraint_functions[ii](design_samples[:,jj])
# return vals
def function(design_samples):
vals = np.empty((design_samples.shape[1]))
for jj in range(design_samples.shape[1]):
vals[jj] = constraints(design_samples[:, jj], [ii])
return vals
X, Y, Z = get_meshgrid_function_data(function, plot_limits, num_pts_1d)
norm = None
cset = axs[ii].contourf(X,
Y,
Z,
levels=np.linspace(Z.min(), Z.max(),
num_contour_levels),
cmap=mpl.cm.coolwarm,
norm=norm)
#for kk in range(len(constraint_functions)):
for kk in range(len(constraints.constraints)):
if ii == kk:
ls = '-'
else:
ls = '--'
axs[kk].contour(X, Y, Z, levels=[0], colors='k', linestyles=ls)
plt.colorbar(cset, ax=axs[ii])
return fig, axs
