def initclusterKA(X, K, distance='euclidean'):
M = X.shape[0]
Dist = spdist.pdist(X, metric=distance) # MxM (condensed)
Dist = spdist.squareform(Dist) # MxM
ResultInd = [0 for i in range(K)]
Xmean = np.mean(X, axis=0) # The mean of all rows of X
Dmean = spdist.cdist(
X, [Xmean]) # Distances between rows of X and the mean of X
# The first centre is the closest point to the mean:
ResultInd[0] = np.argmin(Dmean)
io.updateparallelprogress(K)
for k in range(K - 1):
D = np.min(Dist[:, ResultInd[0:k + 1]], axis=1) # Mx1
C = [0 for m in range(M)] # M points (e.g. genes) # Mx1
for m in range(M):
if m in ResultInd:
continue
tmp = D - Dist[:, m] # Mx1 differences
tmp[tmp < 0] = 0 # All negatives make them zeros
C[m] = np.sum(tmp)
ResultInd[k + 1] = np.argmax(C)
io.updateparallelprogress(K)
Result = X[
ResultInd] # These points are the selected K initial cluster centres
return Result
def mseclusters(X, B, donormalise=True, GDM=None):
Xloc = np.array(X)
Bloc = np.array(B)
if ds.maxDepthOfArray(Xloc) == 2:
Xloc = np.expand_dims(Xloc, axis=0)
Nx = len(Xloc) # Number of datasets
if len(Bloc.shape) == 1:
Bloc = Bloc.reshape(-1, 1)
M = Bloc.shape[0] # Number of genes
K = Bloc.shape[1] # Number of clusters
if GDM is None:
GDMloc = np.ones([Bloc.shape[0], Nx], dtype=bool)
else:
GDMloc = np.array(GDM)
# I commented these two lines after adding GDM
#if any([True if x.shape[0] != M else False for x in Xloc]):
# raise ValueError('Unequal number of genes in datasets and partitions')
mseC = np.zeros([Nx, K], dtype=float)
Nk = [np.sum(b) for b in Bloc.transpose()] # Number of genes per cluster
Nd = [x.shape[1] for x in Xloc] # Number of dimensions per dataset
# Normalise if needed
if donormalise:
Xloc = [pp.normaliseSampleFeatureMat(x, 4) for x in Xloc]
# Calculations
for nx in range(Nx):
reportedprogress = 0
for k in range(K):
# Report progress
if (k - reportedprogress == 100):
io.updateparallelprogress(100)
reportedprogress = k
# WORK
if not any(Bloc[:, k]):
mseC[nx, k] = float('nan')
else:
Xlocloc = Xloc[nx][Bloc[GDMloc[:, nx], k], :]
tmp = nu.subtractaxis(Xlocloc,
np.mean(Xlocloc, axis=0),
axis=0)
tmp = np.sum(np.power(tmp, 2))
mseC[nx, k] = tmp / Nd[nx] / Nk[k]
# Report progress
if (K > reportedprogress):
io.updateparallelprogress(K - reportedprogress)
return np.mean(mseC, axis=0)
def clusterdataset(X, K, methods=None, datasetID=-1):
if methods is None: methods = [['k-means']]
methodsloc = [
n if isinstance(n, (list, tuple, np.ndarray)) else [n] for n in methods
]
# Clustering loop
C = len(methodsloc) # Number of methods
U = [None] * C
for ms in range(C):
if methodsloc[ms][0].lower() in ['k-means', 'kmeans']:
U[ms] = ckmeans(X, K, datasetID, methodsloc[ms][1:])
elif methodsloc[ms][0].lower() in ['hc', 'hierarchical']:
U[ms] = chc(X, K, methodsloc[ms][1:])
io.updateparallelprogress(K * C)
return U
def initclusterKA_memorysaver(X, K, distance='euclidean'):
M = X.shape[0]
#Dist = spdist.pdist(X, metric=distance) # MxM (condensed)
#Dist = spdist.squareform(Dist) # MxM
ResultInd = [0 for i in range(K)]
Xmean = np.mean(X, axis=0) # The mean of all rows of X
Dmean = spdist.cdist(
X, [Xmean],
metric=distance) # Distances between rows of X and the mean of X
# The first centre is the closest point to the mean:
ResultInd[0] = np.argmin(Dmean)
io.updateparallelprogress(K)
for k in range(K - 1):
D = spdist.cdist(
X, X[ResultInd[0:k + 1]], metric=distance
) # (M)x(k+1) Dists of objects to the selected centres
D = np.min(
D, axis=1
) # Mx1: Distances of each of the M objects to its closest already selected centre
C = [0 for m in range(M)] # M objects (e.g. genes) # Mx1
for m in range(M):
if m in ResultInd:
continue
Dists_m = spdist.cdist(
X,
[X[m]
]) # Mx1 distances between all M objects and the m_th object
tmp = D.reshape(M, 1) - Dists_m # Mx1 differences
tmp[tmp < 0] = 0 # All negatives make them zeros
C[m] = np.sum(tmp)
ResultInd[k + 1] = np.argmax(C)
io.updateparallelprogress(K)
Result = X[
ResultInd] # These objects are the selected K initial cluster centres
return Result