def _build_k_prototypes(self, nprot, data):
# K-Means / K-Medoids algorithm
# should return a N-length array with he indices of the chosen data
grouper = K_Means()
grouper._centroid_mode = "index"
grouper.fit(data,
nprot,
iters=300,
mode="cuda",
cuda_mem='manual',
tol=1e-4,
max_iters=300)
centroids = grouper.centroids
nclusters = centroids.shape[0]
# TODO - very inefficient
k_labels = np.zeros(nclusters, dtype=np.int32)
for k in xrange(nclusters):
dist = data - centroids[k]
dist = dist**2
dist = dist.sum(axis=1)
k_labels[k] = dist.argmin()
return k_labels
startE = cuda.event() endE = cuda.event() startE.record() _cu_label_kernel_dists[gridDim,blockDim](dData,dCentroids,dLabels,dDists) endE.record() endE.synchronize() print cuda.event_elapsed_time(startE,endE) startE.record() dDists.copy_to_host(ary = generator._dists) labels = dLabels.copy_to_host(ary = generator.labels) endE.record() endE.synchronize() print cuda.event_elapsed_time(startE,endE) del generator, dData, dCentroids, dLabels, dDists generator = K_Means(mode="numpy") generator.n_clusters = 176 generator.max_iter = 1 generator.fit(data) generator = K_Means(mode="cuda", cuda_mem="manual") generator.n_clusters = 176 generator.max_iter = 1 generator.fit(data) if __name__ = "__main__": main()
