def read_descs(infiles, fmt):
""" read and concatenate matrices from a list of files"""
vl = []
for fname in infiles:
print "reading", fname, "\r",
sys.stdout.flush()
if fmt == 'fvecs':
v = ynumpy.fvecs_read(fname)
elif fmt == 'siftgeo':
v, meta = ynumpy.siftgeo_read(fname)
v = v.astype(numpy.float32)
else: assert False, "unknown format %s" % informat
if v.shape[1] != 0: vl.append(v.T)
return numpy.vstack(vl).T
def read_descs(infiles, fmt):
""" read and concatenate matrices from a list of files"""
vl = []
for fname in infiles:
print "reading", fname, "\r",
sys.stdout.flush()
if fmt == 'fvecs':
v = ynumpy.fvecs_read(fname)
elif fmt == 'siftgeo':
v, meta = ynumpy.siftgeo_read(fname)
v = v.astype(numpy.float32)
else:
assert False, "unknown format %s" % informat
if v.shape[1] != 0: vl.append(v.T)
return numpy.vstack(vl).T
print "queries="
print queries
idx, dis = ynumpy.knn(base, queries, nnn, distance_type = 1)
print "indices="
print idx
print "distances="
print dis
try:
# v, meta = ynumpy.siftgeo_read('/Users/matthijs//Desktop/papers/lhl/trunk/data/test_query_10k.siftgeo')
v, meta = ynumpy.siftgeo_read('/scratch2/bigimbaz/dataset/holidays/siftgeo/hesaff_norm/128300.siftgeo')
v = v.astype('float32')
except Exception, e:
print e
print "generating random data"
v = numpy.random.normal(0, 1, size = (20, 4)).astype(numpy.float32)
v[10:,:] += numpy.tile(numpy.random.uniform(-10, 10, size = (1, 4)),
(10, 1))
else:
print "vectors = "
print v
print "meta info = "
import os
import sys
import numpy as np
sys.path.append('/home/sibo/Documents/Projects/yael/yael_v438')
from yael import ynumpy
# list of available images
image_names = [filename.split('.')[0]
for filename in os.listdir('holidays_100')
if filename.endswith('.jpg')]
# load the SIFTs for these images
image_descs = []
for imname in image_names:
desc, meta = ynumpy.siftgeo_read("holidays_100/%s.siftgeo" % imname)
if desc.size == 0: desc = np.zeros((0, 128), dtype = 'uint8')
# we drop the meta-information (point coordinates, orientation, etc.)
image_descs.append(desc)
# make a big matrix with all image descriptors
all_desc = np.vstack(image_descs)
k = 64
n_sample = k * 1000
# choose n_sample descriptors at random
sample_indices = np.random.choice(all_desc.shape[0], n_sample)
sample = all_desc[sample_indices]
# until now sample was in uint8. Convert to float32
sample = sample.astype('float32')
image_directory = "ukbench_jpg"
sift_directory = "ukbench_siftgeo"
# indices of the images we want to index
image_range = numpy.arange(3000, 3100)
print("Collecting a training set...")
train_set = []
# take descriptors from one image per group from the end of the set
for i in range(10000, 10200, 4):
filename = "%s/ukbench%05d.siftgeo" % (sift_directory, i)
print(" " + filename + "\r")
sys.stdout.flush()
sift_descriptors, geometric_info = ynumpy.siftgeo_read(filename)
train_set.append(sift_descriptors)
train_set = numpy.vstack(train_set)
print("Training set of %d local descriptors in %d dimensions" %
(train_set.shape[0], train_set.shape[1]))
trainset_size = num_gmm_components * 1000
if trainset_size < train_set.shape[0]:
print("Subsampling to %d points" % trainset_size)
subset = numpy.array(
random.sample(range(train_set.shape[0]), trainset_size))
train_set = train_set[subset]
print "queries="
print queries
idx, dis = ynumpy.knn(base, queries, nnn, distance_type=1)
print "indices="
print idx
print "distances="
print dis
try:
# v, meta = ynumpy.siftgeo_read('/Users/matthijs//Desktop/papers/lhl/trunk/data/test_query_10k.siftgeo')
v, meta = ynumpy.siftgeo_read(
'/scratch2/bigimbaz/dataset/holidays/siftgeo/hesaff_norm/128300.siftgeo'
)
v = v.astype('float32')
except Exception, e:
print e
print "generating random data"
v = numpy.random.normal(0, 1, size=(20, 4)).astype(numpy.float32)
v[10:, :] += numpy.tile(numpy.random.uniform(-10, 10, size=(1, 4)),
(10, 1))
else:
print "vectors = "
print v
image_directory = "ukbench_jpg"
sift_directory = "ukbench_siftgeo"
# indices of the images we want to index
image_range = numpy.arange(3000, 3100)
print("Collecting a training set...")
train_set = []
# take descriptors from one image per group from the end of the set
for i in range(10000, 10200, 4):
filename = "%s/ukbench%05d.siftgeo" % (sift_directory, i)
print(" " + filename + "\r")
sys.stdout.flush()
sift_descriptors, geometric_info = ynumpy.siftgeo_read(filename)
train_set.append(sift_descriptors)
train_set = numpy.vstack(train_set)
print("Training set of %d local descriptors in %d dimensions" % (train_set.shape[0], train_set.shape[1]))
trainset_size = num_gmm_components * 1000
if trainset_size < train_set.shape[0]:
print("Subsampling to %d points" % trainset_size)
subset = numpy.array(
random.sample(range(train_set.shape[0]), trainset_size))
train_set = train_set[subset]
print queries
idx, dis = ynumpy.knn(base, queries, nnn, distance_type = 1)
print "indices="
print idx
print "distances="
print dis
try:
# v, meta = ynumpy.siftgeo_read('/Users/matthijs//Desktop/papers/lhl/trunk/data/test_query_10k.siftgeo')
# v, meta = ynumpy.siftgeo_read('/scratch2/bigimbaz/dataset/holidays/siftgeo/hesaff_norm/128300.siftgeo')
v, meta = ynumpy.siftgeo_read('/tmp/128300.siftgeo')
v = v.astype('float32')
except Exception, e:
print e
print "generating random data"
v = numpy.random.normal(0, 1, size = (20, 4)).astype(numpy.float32)
v[10:,:] += numpy.tile(numpy.random.uniform(-10, 10, size = (1, 4)),
(10, 1))
else:
print "vectors = "
print v
print "meta info = "
print meta
