def compare_one_sample_size_full(work_dir, baseline_setup, proposed_setup,
sample_size, result_file):
baseline_dir = os.path.join(work_dir, baseline_setup, sample_size)
proposed_dir = os.path.join(work_dir, proposed_setup, sample_size)
samples = utils.get_immediate_subdirectories(baseline_dir)
count_improve = 0
count_opposite = 0
sample_num = len(samples)
for sample_id in samples:
baseline_sample = os.path.join(baseline_dir, sample_id)
proposed_sample = os.path.join(proposed_dir, sample_id)
sample_improve = compare_one_sample_full(baseline_sample,
proposed_sample, sample_id,
sample_size, result_file)
if sample_improve == '1':
count_improve += 1
elif sample_improve == '0':
count_opposite += 1
result_file.write(
'\nsample_size_result: sample-size=%s, sample-num=%d, improve-num=%d, wrong-num=%d, ratio that %.0f%% statistical significance is made = %.0f%%, wrong ratio = %.0f%%\n'
% (sample_size, sample_num, count_improve, count_opposite,
confidence * 100, float(count_improve) / sample_num * 100,
float(count_opposite) / sample_num * 100))
result_file.write(
'For %d samples in total %d samples, we draw the conclusion the proposed system is better than the baseline with at least %.0f%% statistical significance\n'
% (count_improve, sample_num, confidence * 100))
result_file.close()
def get_annotation_map(data_directory):
"""
:param data_directory:
:type data_directory:
:return:
:rtype:
"""
new_obj_map = {k.strip(): v.strip() for k, v in OBJECT_WNID_MAP.items()}
concerned_objs = [new_obj_map[k] for k in new_obj_map if k in objects]
result = {}
obj_dir_list = get_immediate_subdirectories(data_directory)
for obj_dir in obj_dir_list:
obj = os.path.basename(obj_dir)
if obj not in concerned_objs:
continue
result[obj] = []
anno_dir = os.path.join(obj_dir, 'Annotation')
for anno in get_file_list(anno_dir, format=".xml"):
anno_name = os.path.splitext(os.path.basename(anno))[0]
result[obj].append(anno_name)
return result
def process_openimages(data_dir, class_filter=None):
"""
Process open Images dataset
:param data_dir:
:type data_dir:
:return:
:rtype:
"""
global OPEN_IMAGES_OBJECTS_SET
dataset = dict()
splits = [
os.path.join(data_dir, split)
for split in ['train', 'test', 'validation']
]
for split in splits:
split_dir = os.path.basename(split)
print("Current split:", split_dir)
dataset[split_dir] = {'images': dict(), 'boxes': dict()}
obj_list = get_immediate_subdirectories(split)
for obj in tqdm(obj_list):
obj_name = os.path.basename(obj).lower()
if class_filter:
if obj_name not in class_filter:
continue
img_file_list = get_file_list(obj, format=".jpg")
if len(img_file_list) > 0:
OPEN_IMAGES_OBJECTS_SET.add(obj_name)
label_dir = os.path.join(obj, 'Label')
label_list = get_file_list(label_dir, format=".txt")
for img in img_file_list:
img_name, _ = os.path.splitext(os.path.basename(img))
dataset[split_dir]['images'][img_name] = img
for label in label_list:
label_name, _ = os.path.splitext(os.path.basename(label))
if label_name not in dataset[split_dir]['boxes']:
dataset[split_dir]['boxes'][label_name] = list()
with open(label, 'r') as label_file:
annotations = label_file.readlines()
for annotation in annotations:
dataset[split_dir]['boxes'][label_name].append(
annotation.lower().split())
OPEN_IMAGES_OBJECTS_SET = sorted(OPEN_IMAGES_OBJECTS_SET)
update_label_map(OPEN_IMAGES_OBJECTS_SET)
return dataset
# 将分类标记好bbox的图像一股脑的放在一起，并规范化命名
import sys
import utils
import os
import shutil
import random
if __name__ == "__main__":
root_dir = sys.argv[1]
parent_dir, _ = utils.get_parent_dir(root_dir)
target_dir = os.path.join(parent_dir, 'results')
utils.create_new_empty_dir(target_dir)
os.mkdir(os.path.join(target_dir, 'images'))
os.mkdir(os.path.join(target_dir, 'annotations'))
sub_dirs = utils.get_immediate_subdirectories(root_dir)
total_count = 0
for sub_dir in sub_dirs:
sub_dir = os.path.join(root_dir, sub_dir)
image_dir = os.path.join(sub_dir, 'images')
image_file_list = utils.get_dir_filelist_by_extension(image_dir, 'bmp')
total_count += len(image_file_list)
print('图像总数：%d' % total_count)
file_paths = []
for i in range(1, total_count + 1):
new_name = utils.fixed_length(i, 4)
a = {
'image':
os.path.join(target_dir, 'images', new_name + '.bmp'),
'annotation':
os.path.join(target_dir, 'annotations', new_name + '.xml')
out_layer = TimeDistributed(resnet)(im_layer)
res_squeezed = Reshape(target_shape=(seq_len, 2048))(out_layer)
lstm_out = LSTM(256, return_sequences=True,
activation='sigmoid', recurrent_activation='tanh',
use_bias=True, unit_forget_bias=True)(res_squeezed)
prediction_layer = Dense(2, activation=kb.softmax)(lstm_out)
model = Model(inputs=in_layer, outputs=prediction_layer)
optimizer = adam(lr=0.00001)
model.compile(loss='binary_crossentropy', optimizer=optimizer, metrics=['accuracy'])
print(model.summary())
data_dir = './ActivityDataset'
traj_dir = data_dir + '/' + 'TrajectoriesLong'
train_dir = traj_dir + '/' + 'train'
ex_dirs = get_immediate_subdirectories(train_dir)
acc_arr = []
for i in range(num_iter):
X, Y = get_batch_image(train_dir, batch_size, seq_len)
print(X.shape, Y.shape)
# model.train_on_batch(X, Y)
model.fit(x=X, y=Y, epochs=1, verbose=1)
score, acc = model.evaluate(X, Y, batch_size=batch_size, verbose=1)
acc_arr.append(acc)
print('Step:', i, 'Score: ', score, 'Accuracy:', acc)
if (i % decay_step == 0) and i is not 0:
kb.set_value(optimizer.lr, 0.5 * kb.get_value(optimizer.lr))
if (i % disp_step == 0) and i is not 0:
plt.plot(acc_arr)
plt.pause(0.0005)
'''
count_files = 0
for files in utils.get_immediate_files(folder):
if files.endswith(filetype):
count_files += 1
return count_files
if __name__ == "__main__":
folderName = "downloads"
filesFolderPath = "files"
print("\n")
# Look for folders
dateFolders = utils.get_immediate_subdirectories(folderName)
# Choose a folder
choose_text = "Choose folder to transform files to .pdf:\n"
i = 1
for folder in dateFolders:
choose_text += "(" + str(i) + ") " + dateFolders[i - 1] + "\n"
i += 1
isFolderChooseOk = False
while not isFolderChooseOk:
try:
folder_index = int(input(choose_text))
choose_folder = os.path.join(folderName,
dateFolders[folder_index - 1],