def steinhardt_order(M,L,l=6,rcut=17.5,count=14,verbose=False):
print 'l',l
#Find nearest neighbors for all particles and convert to spherical
#######
#neighbors[i][0]=index of neighbors
#neighbors[i][1]=array[i][x,y,z] of distance between neighbors
########
neighbors=[]
for i in range(M.shape[0]):
N = nn.count_neighbors_index(M,i,L,count=count,rcut=rcut,vectors=True)
if i == 0:
print i, N[1]
neighbors.append([N[0],xyz_to_spherical(N[1])])
if verbose:
print 'neighbors',len(N[0])
#find qlm of each particle
#######
#store as values m=-l-l in nparray
#Qlm[i]=np.array(ql-m:ql+m) for particle i
#Qlm[i]=qlm(i)
######
Qlm=np.zeros((M.shape[0],2*l+1),dtype=complex)
qi=np.zeros((1,2*l+1),dtype=complex)
for i in range(M.shape[0]):
for m in range(-l,l+1):
qi[0][m+l]=qlm(neighbors[i][1],l,m)
Qlm[i][:]=qi
Ql=[]
for i in range(M.shape[0]):
Ql.append(ql(Qlm[i],l))
print sum(Ql)/len(Ql)
return Ql
def single_rotation(M,Z,L,index=0):
theta = []
phi = []
sides = Z.shape[1]/M.shape[1]
v1 = points.vector(M[:,index],Z[:,index*sides],L)
spher = xyz_to_spherical(v1)
theta = spher[:,1]
phi = spher[:,2]
return theta, phi
def rotations(M,Z,L,sides=6):
theta = []
phi = []
zrange = range(0,Z.shape[0],sides)
for i in range(M.shape[0]):
v1 = points.vector(M[k],Z[k],L)
spher = xyz_to_spherical(v1)
theta.append(spher[1])
phi.append(spher[2])
return theta,phi
def single_rotation_box(M,Z,L,index=0):
theta = []
phi = []
sides = Z.shape[1]/M.shape[1]
v = np.zeros((M.shape[0],3))
for k in range(M.shape[0]):
v[k] = points.vector1d(M[k][index],Z[k][index*sides],L[k])
spher = xyz_to_spherical(v)
theta = spher[:,1]
phi = spher[:,2]
return theta, phi
def Qmapq6q4(M,L,count=14,crystal=6,verbose=False,rcut=30):
#Find nearest neighbors for all particles and convert to spherical
#######
#neighbors[i][0]=index of neighbors
#neighbors[i][1]=array[i][x,y,z] of distance between neighbors
########
neighbors=[]
print 'getting neighbors'
for i in range(M.shape[0]):
N = nn.count_neighbors_index(M,i,L,count=count, rcut=rcut,vectors=True)
if len(N[0]) != count:
print 'lenght of N', len(N[0])
neighbors.append([N[0],xyz_to_spherical(N[1])])
#include itself
neighbors[-1][0].append(i)
if verbose:
print 'neighbors',len(N[0])
#find qlm of each particle
#######
#store as values m=-l-l in nparray
#Qlm[i]=np.array(ql-m:ql+m) for particle i
#Qlm[i]=qlm(i)
######
print 'generating qlm matrix'
l=4
qlm_matrix=np.zeros((M.shape[0],2*l+1),dtype=complex)
for i in range(M.shape[0]):
for m in range(-l,l+1):
qlm_matrix[i][m+l]=qlm(neighbors[i][1],l,m)
Ql4=np.zeros((M.shape[0]),dtype=complex)
print 'generating Ql_avg'
for i in range(M.shape[0]):
qlm_avg=np.zeros((2*l+1),dtype=complex)
for m in range(-l,l+1):
for k in neighbors[i][0]:
qlm_avg[m+l] += qlm_matrix[k,m+l]
qlm_avg[m+l]/=len(neighbors[i][0])
Ql4[i]=ql(qlm_avg,l)
print 'Ql4_avg=',l,Ql4.sum()/Ql4.shape[0]
Ql6=np.zeros((M.shape[0]),dtype=complex)
l=6
qlm_matrix=np.zeros((M.shape[0],2*l+1),dtype=complex)
for i in range(M.shape[0]):
for m in range(-l,l+1):
qlm_matrix[i][m+l]=qlm(neighbors[i][1],l,m)
for i in range(M.shape[0]):
qlm_avg=np.zeros((2*l+1),dtype=complex)
for m in range(-l,l+1):
for k in neighbors[i][0]:
qlm_avg[m+l]+=qlm_matrix[k,m+l]
qlm_avg[m+l]/=len(neighbors[i][0])
Ql6[i]=ql(qlm_avg,l)
print 'Ql6_avg=',l,Ql6.sum()/Ql6.shape[0]
return Ql4,Ql6
def solid_particles(M,L,count=14,l=6,c_cut=0.15,crystal=6,verbose=False,rcut=30):
#Find nearest neighbors for all particles and convert to spherical
#######
#neighbors[i][0]=index of neighbors
#neighbors[i][1]=array[i][x,y,z] of distance between neighbors
########
neighbors=[]
for i in range(M.shape[0]):
N = nn.count_neighbors_index(M,i,L,count=count,rcut=rcut, vectors=True)
neighbors.append([N[0],xyz_to_spherical(N[1])])
if verbose:
print 'neighbors',len(N[0])
#find qlm of each particle
#######
#store as values m=-l-l in nparray
#Qlm[i]=np.array(ql-m:ql+m) for particle i
#Qlm[i]=qlm(i)
######
Qlm=np.zeros((M.shape[0],2*l+1),dtype=complex)
qi=np.zeros((1,2*l+1),dtype=complex)
for i in range(M.shape[0]):
for m in range(-l,l+1):
qi[0][m+l]=qlm(neighbors[i][1],l,m)
Qlm[i][:]=qi
#for nearest neighbors find the Sij vector
S=[]
for i in range(M.shape[0]):
S.append(Sij(Qlm[i],Qlm[neighbors[i][0]]))
#for each particle lets counts its neighbors
#if it has more than ccut we say it is a solid
iscrystal=[0.0 for i in range(M.shape[0])]
counter=0
store_j = []
for i in S:
#if verbose:
# print i
count=0
for j in i:
#this is the cutoff for declaring 2 particles connected
if j.real>c_cut:
store_j.append(j.real)
count+=1
if count >= crystal:
iscrystal[counter]=1
counter+=1
try:
print sum(store_j)/len(store_j)
except:
print 'no bonds'
num=iscrystal.count(1)
print 'crystals',num
return num, iscrystal
def Qmap(M,L,count=14,l=6,crystal=6,verbose=False,rcut=30):
#Find nearest neighbors for all particles and convert to spherical
#######
#neighbors[i][0]=index of neighbors
#neighbors[i][1]=array[i][x,y,z] of distance between neighbors
########
neighbors=[]
#otree = node(L,delta=(M.shape[0]/2)**0.333)
#otree.add_particles(M)
print 'getting neighbors'
for i in range(M.shape[0]):
#N = cn_tree_index(M,i,otree,L,count=count,rcut=rcut)
N = nn.count_neighbors_index(M,i,L,count=count,rcut=rcut,vectors=True)
neighbors.append([N[0],xyz_to_spherical(N[1])])
#include itself
neighbors[-1][0].append(i)
if verbose:
print 'neighbors',len(N[0])
#find qlm of each particle
#######
#store as values m=-l-l in nparray
#Qlm[i]=np.array(ql-m:ql+m) for particle i
#Qlm[i]=qlm(i)
######
print 'generating qlm matrix'
qlm_matrix=np.zeros((M.shape[0],2*l+1),dtype=complex)
for i in range(M.shape[0]):
for m in range(-l,l+1):
qlm_matrix[i][m+l]=qlm(neighbors[i][1],l,m)
Ql=np.zeros((M.shape[0]),dtype=complex)
print 'generating Ql_avg'
for i in range(M.shape[0]):
qlm_avg=np.zeros((2*l+1),dtype=complex)
for m in range(-l,l+1):
for k in neighbors[i][0]:
qlm_avg[m+l]+=qlm_matrix[k,m+l]
qlm_avg[m+l]/=len(neighbors[i][0])
Ql[i]=ql(qlm_avg,l)
print 'Ql=',l,Ql.sum()/Ql.shape[0]
return Ql