def main():
with open(TRAIN_FILENAME) as train_data_file:
image_ids_whale_ids, whale_ids = utility.read_train(train_data_file)
images_reader = ImagesReader(IMAGES_DIR)
images_reader.pre_process(image_processors.region_crop_gray_downscale, rewrite=False, threads=1)
all_train_images_ids = image_ids_whale_ids[:, 0]
unique_train_images_ids = set(all_train_images_ids)
all_images_ids = set(images_reader.image_ids)
result_images_ids = all_images_ids.difference(unique_train_images_ids)
train_image_id_whale_id = dict(image_ids_whale_ids)
print('Reading train data\n')
x_train = np.asarray([images_reader.read_image_vector(image_id)
for image_id in all_train_images_ids])
y_train = np.asarray([train_image_id_whale_id[image_id] for image_id in all_train_images_ids])
features_cnt = len(x_train[0])
num_targets = len(set(y_train))
clf = simple_cnn.CNN(features_cnt, num_targets,
num_epochs=10,
fresh_start=False,
dump_dir="network_weights/",
filename_to_dump="net.w")
# clf = SVC(probability=True)
print('Fitting\n')
clf.fit(x_train, y_train)
print('Reading test data\n')
x_test = np.array([images_reader.read_image_vector(image_id)
for image_id in result_images_ids])
print('Predicting\n')
y_predicted = clf.predict_proba(x_test)
print('Writing submission')
with open(SUBMISSION_FILENAME, 'w') as submission_file:
utility.write_submission(whale_ids, result_images_ids, y_predicted, submission_file)
return 0
def main():
"""
Split whale images by groups according to their IDs in train.CSV
"""
with open(TRAIN_CSV) as inp:
image_ids_whale_ids, _ = utility.read_train(inp)
if not os.path.exists(GROUPS):
try:
os.makedirs(GROUPS)
except:
print("Something goes wrong. Please, check groups dir was created and you can copy a file to it!")
for name_id, wid in image_ids_whale_ids:
name = "w_%s.jpg" % name_id
print(name, wid)
target_dir = os.path.join(GROUPS, str(wid))
if not os.path.exists(target_dir):
os.makedirs(target_dir)
shutil.copy(os.path.join(PREPROCESSING_DIR, name), os.path.join(target_dir, name))