def test_point_correctness():
import itertools
stencil = [-1, 0, 1]
ndim = 3
n = 2000
stencil = itertools.product(*[stencil] * ndim)
stencil = np.array(list(stencil)).astype(np.int32)
points = (np.random.rand(n, ndim) * [1, 2, 3]).astype(np.float32)
scale = 0.1
spec = GridSpec(points, float(scale))
offsets = spec.stencil(stencil).astype(np.int32)
grid = PointGrid(spec, points, offsets)
pairs = grid.pairs()
from scipy.spatial import cKDTree
tree = cKDTree(points)
tree_pairs = tree.query_pairs(scale, output_type='ndarray')
print(tree_pairs)
print(pairs)
assert np.alltrue(
npi.unique(tree_pairs) == npi.unique(np.sort(pairs, axis=1)))
def create_pre_compute():
with open('./Data/Trajectories.txt', 'r') as f:
# trajectory_num = []
trajectories_list = []
for line in f:
lst = re.split(r'[ ]', line)
lst.pop(-1)
tmp = [float(i) for i in lst]
# trajectory_num.append(tmp.pop(-1))
trajectories_list.append(tmp)
trajectories = np.zeros([len(trajectories_list), 24])
for i, trajectory in enumerate(trajectories_list):
for j, stop_area in enumerate(trajectory):
trajectories[i, j] = int(stop_area)
unique_trajectories = npi.unique(trajectories)
label_uniques = npi.indices(unique_trajectories, trajectories)
np.save('./Data/label_uniques.npy', label_uniques)
np.save('./Data/trajectories_uniques.npy', unique_trajectories)
np.save('./Data/np_trajectories.npy', trajectories)
# trajectories = np.load('./Data/np_trajectories.npy')
# unique_trajectories = np.load('./Data/trajectories_uniques.npy')
# label_uniques = np.load('./Data/label_uniques.npy')
dist_pre_compute = np.zeros([len(unique_trajectories), len(unique_trajectories)])
for i in range(0, len(unique_trajectories)-1):
for j in range(i+1, len(unique_trajectories)):
if j % 2000 == 0:
print([i, j])
dist_pre_compute[i, j] = dist_pre_compute[j, i] = levenshtein(unique_trajectories[i, :], unique_trajectories[j, :])
# dist_pre_compute[i, j] = dist_pre_compute[j, i] = dtw(unique_trajectories[i, :], unique_trajectories[j, :])
np.save('./Data/pre_compute_dtw', dist_pre_compute)
def find_squares(img):
yuv = cv.cvtColor(img, cv.COLOR_BGR2YUV)
y, u, v = cv.split(yuv)
#img = cv.GaussianBlur(img, (5, 5), 0)
squares = []
#for gray in cv.split(img):
_retval, bin = cv.threshold(y, 0, 255, cv.THRESH_OTSU)
binn = cv.bitwise_not(bin)
c = cv.getStructuringElement(cv.MORPH_ELLIPSE, (7, 7))
opened = cv.morphologyEx(binn, cv.MORPH_OPEN, c)
bin, contours, _hierarchy = cv.findContours(opened, cv.RETR_LIST,
cv.CHAIN_APPROX_SIMPLE)
contours = sorted(contours, key=cv.contourArea, reverse=True)
topN = 3
maxLoop = topN if len(contours) >= topN else len(contours)
for i in range(0, maxLoop):
cnt = contours[i]
cnt_len = cv.arcLength(cnt, True)
cnt = cv.approxPolyDP(cnt, 0.02 * cnt_len, True)
if len(cnt) == 4 and cv.contourArea(cnt) > 1000 and cv.isContourConvex(
cnt):
cnt = cnt.reshape(-1, 2)
max_cos = np.max([
angle_cos(cnt[i], cnt[(i + 1) % 4], cnt[(i + 2) % 4])
for i in list(range(4))
])
if max_cos < 0.1:
squares.append(cnt)
squares = npi.unique(squares)
return squares
def merge(self, other):
vertices = np.concatenate([self.vertices, other.vertices], axis=0)
faces = np.concatenate([self.faces, other.faces + len(self.vertices)],
axis=0)
_, _idx, _inv = npi.unique(vertices,
return_index=True,
return_inverse=True)
return type(self)(vertices[_idx], _inv[faces])
def compute_face_incidence(self):
unsorted_edges = self.edges().reshape(-1, 2)
sorted_edges = np.sort(unsorted_edges, axis=-1)
unique_edges, edge_indices = npi.unique(sorted_edges, return_inverse=True)
face_indices = np.arange(self.faces.size) // 3
orientation = sorted_edges[:, 0] == unsorted_edges[:, 0]
incidence = scipy.sparse.csr_matrix((orientation * 2 - 1, (edge_indices, face_indices)))
return incidence, unique_edges
def create_np():
names_dict, label_list = points_dict()
for file_tmp in os.listdir("./Data/CoordinatesInput"):
if file_tmp.endswith(".txt"):
str_file = file_tmp.title()
str_file_pre = './Data/CoordinatesInput/%s.txt' % (str_file[:-4])
str_file_np = './Data/npData/%s' % (str_file[:-4])
with open(str_file_pre, 'r') as f:
print(str_file)
all_coordinates = []
for line in f:
line = line.strip('\n')
tmp = ''
coordinates_list = []
for sign in line:
if sign != '[' and sign != ']' and sign != "'" and sign != '"' and sign != ' ':
if sign == ',':
coordinates_list.append(tmp)
tmp = ''
else:
tmp += sign
coordinates_list.append(tmp)
all_coordinates.append(coordinates_list)
if len(all_coordinates) > 0:
data = np.zeros([len(all_coordinates), 6])
for idx, item1 in enumerate(all_coordinates):
for idy, item2 in enumerate(item1):
data[idx][idy] = float(item2)
tmp = data[:, :2]
unique_coordinate = npi.unique(tmp)
label = npi.indices(unique_coordinate, tmp)
max_cluster = np.max(label) + 1
count = np.zeros(max_cluster)
for i in label:
count[i] += 1
temp = count
idc = np.zeros(max_cluster)
for i in range(max_cluster):
idx = np.argmax(temp)
idc[idx] = max_cluster - (i + 1)
temp[idx] = 0
for idx, count_label in enumerate(idc):
if count_label < max_cluster - math.floor(max_cluster * 0.1):
label[np.where(label == idx)] = max_cluster
j = 0
for i in range(max_cluster + 1):
tmp = np.where(label == i)
if len(tmp[0]) > 0:
data[np.where(label == i), 5] = label_list[names_dict[str_file[:-4]]][j]
j += 1
np.save(str_file_np, data)
def load_stl(filename):
dtype = [('normal', '<f4', 3,),('vertex', '<f4', (3,3)), ('abc', '<u2', 1,)]
with open(filename, 'rb') as fh:
header = np.fromfile(fh, '<c', 80)
triangles = np.fromfile(fh, '<u4', 1)[0]
data = np.fromfile(fh, dtype, triangles)
vertices, triangles = npi.unique(data['vertex'].reshape(-1, 3), return_inverse=True)
return Mesh(vertices, triangles.reshape(-1, 3))
def compute_vertex_incidence(self):
unsorted_edges = self.edges().reshape(-1, 2)
sorted_edges = np.sort(unsorted_edges, axis=-1)
vertex_indices = npi.unique(sorted_edges)
edge_indices = np.arange(vertex_indices.size) // 2
orientation = vertex_indices == vertex_indices[:, 0:1]
incidence = scipy.sparse.csr_matrix(
((orientation * 2 - 1).flatten(), (edge_indices, vertex_indices.flatten())))
return incidence
def compute_face_incidence(self):
unsorted_edges = self.edges().reshape(-1, 2)
sorted_edges = np.sort(unsorted_edges, axis=-1)
unique_edges, edge_indices = npi.unique(sorted_edges,
return_inverse=True)
face_indices = np.arange(self.faces.size) // 3
orientation = sorted_edges[:, 0] == unsorted_edges[:, 0]
incidence = scipy.sparse.csr_matrix(
(orientation * 2 - 1, (edge_indices, face_indices)))
return incidence, unique_edges
def minimize_objective(
x, y, cv_splits, dim,
dimension_bounds): # Possible to leave out and integrate in Max_EI
temp = npi.unique(x, return_index=1)
x = temp[0]
y = y[temp[1]]
model_weights = build_ensemble(x, y, cv_splits,
sklearn.metrics.mean_absolute_error)
trained_models = train_models(x, y)
new_points_prediction = max_EI(trained_models, model_weights, x,
y.reshape(-1, 1), dim, dimension_bounds)
return ((trained_models, model_weights, new_points_prediction))
def compute_vertex_incidence(self):
unsorted_edges = self.edges().reshape(-1, 2)
sorted_edges = np.sort(unsorted_edges, axis=-1)
vertex_indices = npi.unique(sorted_edges)
edge_indices = np.arange(vertex_indices.size) // 2
orientation = vertex_indices == vertex_indices[:, 0:1]
incidence = scipy.sparse.csr_matrix(
((orientation * 2 - 1).flatten(), (edge_indices,
vertex_indices.flatten())))
return incidence
def test_point_correctness():
import itertools
stencil = [-1, 0, 1]
ndim = 3
n = 2000
stencil = itertools.product(*[stencil]*ndim)
stencil = np.array(list(stencil)).astype(np.int32)
points = (np.random.rand(n, ndim) * [1, 2, 3]).astype(np.float32)
scale = 0.1
spec = GridSpec(points, float(scale))
offsets = spec.stencil(stencil).astype(np.int32)
grid = PointGrid(spec, points, offsets)
pairs = grid.pairs()
from scipy.spatial import cKDTree
tree = cKDTree(points)
tree_pairs = tree.query_pairs(scale, output_type='ndarray')
print(tree_pairs)
print(pairs)
assert np.alltrue(npi.unique(tree_pairs) == npi.unique(np.sort(pairs, axis=1)))
def remove_disconnected_elems_from_mesh(mesh, split_facets, C_elems):
"""Elements which are completely disconnected from the rest of the mesh can safely be removed"""
components = list(nx.connected_components(C_elems))
print "Number of components after facet disconnection = ", len(components)
if len(components) == 1:
#Everything connected so nothing to be done
return mesh, split_facets, C_elems
#Some elems are disconnected by split facets so remove them from the mesh
lc_num = np.argmax([len(c) for c in components]) #Largest component
connected_elems = np.array(list(components[lc_num]))
isolated_elems = np.setdiff1d(range(len(mesh.elems)), connected_elems)
print "Number elements removed = ", len(isolated_elems)
assert isolated_elems.any()
#update split facets
removed_facets = npi.unique(
np.sort(np.sort(np.array([
np.roll(mesh.elems[isolated_elems], i, axis=1).flatten()
for i in range(mesh.edim - 1)
]).transpose(),
axis=1),
axis=1))
split_facets = split_facets[np.logical_not(
npi.in_(split_facets, np.array(removed_facets)))]
#Update the connectivity graph
C_elems.remove_nodes_from(isolated_elems)
ediff = np.zeros(len(mesh.elems), dtype=int)
for enum in isolated_elems:
ediff[enum:] += 1
elem_connections = np.array(C_elems.edges())
elem_connections -= ediff[elem_connections]
C_elems = nx.Graph(elem_connections.tolist())
# #Renumber and update mesh
is_elem_removed = np.zeros(len(mesh.elems))
is_elem_removed[isolated_elems] = 1
mesh, vdiff = renumber_mesh(mesh, is_elem_removed, return_vdiff=True)
#Renumber the split vertex numbers
split_facets -= vdiff[split_facets]
return mesh, split_facets, C_elems
def validate(self, *, population: np.ndarray, **kwargs) -> np.ndarray:
"""Removes duplicate individuals from population
Args:
population (np.ndarray): the population to validate
**kwargs: keyword arguments for plugins
Returns:
np.ndarray: same width as population, likely has less rows
"""
# the first part eliminates individuals with duplicate genes
# the second part eliminates duplicate individuals
population_sorted = np.sort(population, axis=-1)
population = population[(population_sorted[..., 1:] !=
population_sorted[..., :-1]).all(-1)]
return unique(np.sort(population, axis=1))
def generate_vertices(self, group):
"""instantiate a full sphere by repeating the transformed fundamental domain
Returns
-------
ndarray, [n, 3], float
all points in the geometry, on a unit sphere
"""
points = np.empty((group.index, group.order, self.topology.P0, 3), np.float)
PP = self.decomposed
for i, B in enumerate(group.basis):
for t, b in enumerate(B.reshape(-1, 3, 3)):
b = util.normalize(b.T).T # now every row is a normalized vertex
P = np.dot(b, PP.T).T # go from decomposed coords to local coordinate system
points[i, t] = P
# make single unique point list
return npi.unique(points.reshape(-1, 3))
def load_stl(filename):
dtype = [(
'normal',
'<f4',
3,
), ('vertex', '<f4', (3, 3)), (
'abc',
'<u2',
1,
)]
with open(filename, 'rb') as fh:
header = np.fromfile(fh, '<c', 80)
triangles = np.fromfile(fh, '<u4', 1)[0]
data = np.fromfile(fh, dtype, triangles)
vertices, triangles = npi.unique(data['vertex'].reshape(-1, 3),
return_inverse=True)
return Mesh(vertices, triangles.reshape(-1, 3))
def refine_sphere(sphere):
"""given a spherical mesh, insert a new vertex on every edge
Parameters
----------
sphere : skcg.Mesh instance
Returns
-------
skcg.Mesh instance
"""
vertices = sphere.vertices
edges = npi.unique(sphere.ordered_edges())
new_vertices = vertices[edges].mean(axis=1)
new_vertices /= np.linalg.norm(new_vertices, axis=1, keepdims=True)
sphere = triangulate_convex(np.concatenate((vertices, new_vertices)))
direction = collision.mymath.dot(sphere.face_normals(), sphere.face_centroids()) > 0
faces = np.where(direction[:, None], sphere.faces[:, ::+1], sphere.faces[:, ::-1])
sphere = Mesh(sphere.vertices, faces)
return sphere
def refine_sphere(sphere):
"""given a spherical mesh, insert a new vertex on every edge
Parameters
----------
sphere : skcg.Mesh instance
Returns
-------
skcg.Mesh instance
"""
vertices = sphere.vertices
edges = npi.unique(sphere.ordered_edges())
new_vertices = vertices[edges].mean(axis=1)
new_vertices /= np.linalg.norm(new_vertices, axis=1, keepdims=True)
sphere = triangulate_convex(np.concatenate((vertices, new_vertices)))
direction = collision.mymath.dot(sphere.face_normals(),
sphere.face_centroids()) > 0
faces = np.where(direction[:, None], sphere.faces[:, ::+1],
sphere.faces[:, ::-1])
sphere = Mesh(sphere.vertices, faces)
return sphere
def list_equivalent_rules(rules, preferences='none'):
"""
:param rules: dictionary containing rule_names (key) and rule_values (boolean array)
:return: list of lists. each inner lists contains the rule_names with the same rule_value, ordered
"""
rule_names, rule_values = zip(*rules.items())
rule_values = np.vstack(rule_values)
equivalent_values, duplicate_idx = npi.unique(rule_values,
return_inverse=True)
n_equivalent = len(equivalent_values)
# create list of rules containing equivalence classes
if preferences == 'none':
equivalent_rules = [[]] * n_equivalent
for j in range(n_equivalent):
rule_idx = np.flatnonzero(j == duplicate_idx)
equivalent_rules[j] = [rule_names[k] for k in rule_idx]
else:
# order rules from first to last
raise NotImplementedError()
return equivalent_rules, equivalent_values
def get_facets(self):
facets = np.vstack([
np.roll(self.elems, i, axis=1).flatten()
for i in range(self.edim - 1)
]).transpose()
return npi.unique(np.sort(facets, axis=1))
def unique(ar,
return_index=False,
return_inverse=False,
return_counts=False):
return numpy_indexed.unique(ar, None, return_index, return_inverse,
return_counts)
# Drop all triangles with vertex on one side of split line
keep = []
for tri in fileMesh.vectors:
if not any([x[minDim]>=split for x in tri]):
keep.append(tri)
keep = np.array(keep)
fileMesh.vectors = keep
#plotMesh(fileMesh)
goodDir1 = (minDim+1)%3
goodDir2 = (minDim+2)%3
flatTriangles = keep[:,:,[goodDir1,goodDir2]]
# Drop lines that appear in two triangles - part of interior of part
# Sort the 2 points in each edge to ensure collision. Use 2 sort dirs to break ties
edges = [np.array([sorted([x[0,:],x[1,:]], key=lambda v: v[0]+0.0001*v[1]),
sorted([x[1,:],x[2,:]], key=lambda v: v[0]+0.0001*v[1]),
sorted([x[2,:],x[0,:]], key=lambda v: v[0]+0.0001*v[1])]) for x in flatTriangles]
edges = np.reshape(edges,(-1,4))
edges,counts = npi.unique(np.around(edges,decimals=point_accuracy),return_count=True)
edges = np.reshape(edges[counts==1,:],(-1,2,2))
# Write to svg
svg = svgwrite.Drawing(fName.split('.stl')[0]+'.svg', profile='full',size=('1000mm', '1000mm'), viewBox=('0 0 1000 1000'))
for e in edges:
svg.add(svg.line(tuple(e[0,:].tolist()),tuple(e[1,:].tolist()), stroke=svgwrite.rgb(0,0,0,'%')))
svg.save()
print "Saved " + str(edges.shape[0]) + " edges."
def merge(self, other):
vertices = np.concatenate([self.vertices, other.vertices], axis=0)
faces = np.concatenate([self.faces, other.faces + len(self.vertices)], axis=0)
_, _idx, _inv = npi.unique(vertices, return_index=True, return_inverse=True)
return type(self)(vertices[_idx], _inv[faces])
