def train_sample(feature_str, label, pos_train=0.5, neg_train=1000):
"""Perform training and testing using disjont samples from the full
set of label_values. This is equivalent to doing one round of cross
validation (see classipy.cross_validation) only it keeps the values around
for display.
Args:
"""
all_hashes = list(cass.get_image_hashes())
pos_hashes = [_[0] for _ in cass.buffered_get_row(cf_labels, label)]
neg_hashes = list(set(all_hashes) - set(pos_hashes))
if 0 < pos_train <= 1: pos_train = int(pos_train * len(pos_hashes))
if 0 < neg_train <= 1: neg_train = int(neg_train * len(neg_hashes))
# Choose a training sample and a testing sample
if len(pos_hashes) < pos_train:
raise ValueError('Not enough positive examples %s(%d)' % \
(label, len(pos_hashes)))
if len(neg_hashes) < neg_train:
raise ValueError('Not enough negative examples %s(%d)' % \
(label, len(neg_hashes)))
pos_sample = random.sample(pos_hashes, pos_train)
neg_sample = random.sample(neg_hashes, neg_train)
labels = [-1 for _ in neg_sample] + [1 for _ in pos_sample]
values = cass.get_feature_values(feature_str, neg_sample+pos_sample)
global label_values
label_values = zip(labels, values)
print 'Training classifier with sample %d' % len(label_values)
train_classifier(label_values)
def split_train_test(label, pos_train=0.5, neg_train=1000):
"""Splits the data into a training and testing sample.
Arguments:
pos_train:
0 < pos_train <= 1.0: the portion of the set to use for training
1 < pos_train: the number of images to use for training
Returns:
(train, test): Training and testing set, where
train: Iterable of tuples for a training set
[(-1|1, hash), ...]
test: Iterable of tuples for a testing set
[(-1|1, hash), ...]
"""
all_hashes = set(cass.get_image_hashes())
pos_hashes = [_[0] for _ in cass.buffered_get_row(cf_labels, label)]
neg_hashes = set(all_hashes) - set(pos_hashes)
if 0 < pos_train <= 1:
pos_train = int(pos_train * len(pos_hashes))
if 0 < neg_train <= 1:
neg_train = int(neg_train * len(neg_hashes))
pos_sample = random.sample(pos_hashes, pos_train)
neg_sample = random.sample(neg_hashes, neg_train)
# Sample from pos and neg
train = [(-1, k) for k in neg_sample] + \
[(1, k) for k in pos_sample]
neg_sample = random.sample(neg_hashes, len(pos_hashes))
# Use all of the images for testing for now
test = [(-1, k) for k in neg_sample] + \
[(1, k) for k in pos_hashes]
print 'Train: %d Test: %d' % (len(train), len(test))
return train, test
def build_label_index(buffer_size=1000):
# Get label
def reverse(items):
for k, v in items:
d = json.loads(v)
g = re.match('(\w+)-(?:\d+)\.jpg', d['filename'])
if g: yield g.groups()[0], {k:''}
items = cass.buffered_get_row(cass.cf_images, 'image_metadata')
items = reverse(items)
# Insert into cass by chunks
count = 0
while items:
chunk = list(itertools.islice(items, buffer_size))
for row, column in chunk:
cf_labels.insert(row, column)
count += len(chunk)
if not chunk: break
print("Built index with %d items" % count)
