RocketGit

hahattpro / crowd_counting_framework (public) (License: Unspecified) (since 2020-12-23) (hash sha1)

backupfrom https://github.com/ttpro1995/crowd_counting_framework

Clone URLs: https://rocketgit.com/user/hahattpro/crowd_counting_framework ssh://rocketgit@ssh.rocketgit.com/user/hahattpro/crowd_counting_framework git://git.rocketgit.com/user/hahattpro/crowd_counting_framework

List of commits:

Subject	Hash	Author	Date (UTC)
demo	3d5fa0d764843d4fd780e44547fd0b13a367c6bc	Thai Thien	2020-11-27 19:28:37
notable image	b0f2b0e9732c2f2031831eb80eeaf7e6dab1f5bd	Thai Thien	2020-11-22 07:11:16
notebook	5b8b840bc1fd0d806a13646477988534f4982fac	Thai Thien	2020-11-22 07:10:42
l2_adamw2_bigtail13i_t13_sha	7fd22510845dec808100058190a202b805bf6a02	Thai Thien	2020-09-23 14:17:00
l2_adamw2_bigtail13i_t12_sha	bf67476db55f55f348068bc7683f68834659cc96	Thai Thien	2020-09-23 14:14:35
l2_adamw2_bigtail13i_t11_sha	1365b305bf541a3d941d173b169fa14aa811beb1	Thai Thien	2020-09-22 19:31:17
l2_adamw2_bigtail13i_t10_sha	4f6f2b8d3ac4654cc2ed94ca6900643f3ca07be9	Thai Thien	2020-09-22 19:27:24
l2_adamw2_bigtail13i_t9_sha	576b424688dff640a0a7d448cbd615cfb4f33d54	Thai Thien	2020-09-22 19:21:17
t8	9e00ee3a50c93bbd726ebd8554a68f535d836691	Thai Thien	2020-09-22 19:11:37
split a	a7b2115bee8ab76497528b84aee09c79631575b1	Thai Thien	2020-09-22 17:48:37
load_data_shanghaitech_256_v2 and l2_adamw2_bigtail13i_t7_sha	07bca1a460c30ed7ec7269d72dad99eef6fc96b3	Thai Thien	2020-09-22 17:04:45
adamw1_ccnnv7_t7_jhu	c5f106103aa3696827627f5e815105d9432e2acb	Thai Thien	2020-09-16 18:21:22
adamw1_ccnnv7_t6_jhu.sh	5c8644da7445b3dd4f6d2611d564592e8e8ed45c	Thai Thien	2020-09-16 18:19:07
adamw1_CompactCNNV7_t6_jhu	1fa1d544a254cd042bcdd36e8de2984fe3d5244f	Thai Thien	2020-09-16 17:56:15
adamw1_CompactCNNV7_t5_jhu	40720162c9e9b731fb852ed7e3e191228e421ced	Thai Thien	2020-09-16 17:46:20
test_data cache	73633cbd6d70448268a3e1534440601ddcf75276	Thai Thien	2020-09-16 17:12:57
fix test data not exist	474f506204251e26825801499f6bc503f44f8410	Thai Thien	2020-09-16 16:41:00
fix test data not exist	6377e18d87e0fabbf9b4ed143e26ae5912c0b872	Thai Thien	2020-09-16 16:40:06
jhucrowd_downsample_512	30b8b9a63e93a5631f003858ef2ab63dc62c6563	Thai Thien	2020-09-16 16:17:02
fix	77d74f09874a7be0d8fcec141597be4ce331bf0c	Thai Thien	2020-09-16 16:08:39

Commit 3d5fa0d764843d4fd780e44547fd0b13a367c6bc - demo
Author: Thai Thien
Author date (UTC): 2020-11-27 19:28
Committer name: Thai Thien
Committer date (UTC): 2020-11-27 19:28
Parent(s): b0f2b0e9732c2f2031831eb80eeaf7e6dab1f5bd
Signing key:
Tree: d45ac55223b73476c60745e63c1036b80fe15c26

File	Lines added	Lines deleted
args_util.py	1	1
data_flow.py	83	0
debug/perfomance_test_on_shb.py	3	0
demo_app/__init__.py	0	0
demo_app/predict_only.py	15	0

File args_util.py changed (mode: 100644) (index 77806a9..d68650a)
...	...	def meow_parse():
136	136	# help="if true, use mse and negative ssim as loss function")	# help="if true, use mse and negative ssim as loss function")
137	137	parser.add_argument('--loss_fn', action="store", default="MSE", type=str)	parser.add_argument('--loss_fn', action="store", default="MSE", type=str)
138	138	parser.add_argument('--optim', action="store", default="adam", type=str)	parser.add_argument('--optim', action="store", default="adam", type=str)
139		parser.add_argument('--eval_only', action="store_true", default=False)
	139		parser.add_argument('--eval_only', action="store_true", default=False, help="only evaluate no train")
140	140	arg = parser.parse_args()	arg = parser.parse_args()
141	141	return arg	return arg
142	142

File data_flow.py changed (mode: 100644) (index 44b669e..029f79b)
...	...	def get_dataloader(train_list, val_list, test_list, dataset_name="shanghaitech",
1451	1451	return train_loader, train_loader_for_eval, val_loader, test_loader	return train_loader, train_loader_for_eval, val_loader, test_loader
1452	1452	else:	else:
1453	1453	return train_loader, val_loader, test_loader	return train_loader, val_loader, test_loader
	1454
	1455
	1456		def simple_predict_data_load_fn(img_path):
	1457		img_name = img_path.split("/")[-1]
	1458		# when debug, give information on p_count and img_name
	1459		debug_info = {"img_path":img_path,
	1460		"name": img_name}
	1461		img_origin = Image.open(img_path).convert('RGB')
	1462		return img_origin, debug_info
	1463
	1464
	1465		class PredictListDataset(Dataset):
	1466		def __init__(self, root, shape=None, shuffle=True, transform=None, batch_size=1,
	1467		debug=False,
	1468		num_workers=0):
	1469		"""
	1470		if you have different image size, then batch_size must be 1
	1471		:param root:
	1472		:param shape:
	1473		:param shuffle:
	1474		:param transform:
	1475		:param train:
	1476		:param debug: will print path of image
	1477		:param seen:
	1478		:param batch_size:
	1479		:param num_workers:
	1480		"""
	1481
	1482		if shuffle:
	1483		random.shuffle(root)
	1484
	1485		self.nSamples = len(root)
	1486		self.lines = root
	1487		self.transform = transform
	1488
	1489		self.cache_train = {}
	1490		self.cache_eval = {}
	1491		self.debug = debug
	1492		self.shape = shape
	1493
	1494		self.batch_size = batch_size
	1495		self.num_workers = num_workers
	1496
	1497		def __len__(self):
	1498		return self.nSamples
	1499
	1500		def __getitem__(self, index):
	1501		assert index <= len(self), 'index range error'
	1502		img_path = self.lines[index]
	1503		# if self.debug:
	1504		# print(img_path)
	1505		# try to check cache item if exist
	1506		img, info = simple_predict_data_load_fn(img_path)
	1507		if self.transform is not None:
	1508		if isinstance(img, list):
	1509		# for case of generate multiple augmentation per sample
	1510		img_r = [self.transform(img_item) for img_item in img]
	1511		img = img_r
	1512		else:
	1513		img = self.transform(img)
	1514		return img, info
	1515
	1516
	1517		def get_predict_dataloader(data_list, visualize_mode=False, batch_size=1,
	1518		debug=False, test_size=1):
	1519
	1520		if visualize_mode:
	1521		transformer = transforms.Compose([
	1522		transforms.ToTensor()
	1523		])
	1524		else:
	1525		transformer = transforms.Compose([
	1526		transforms.ToTensor(), transforms.Normalize(mean=[0.485, 0.456, 0.406],
	1527		std=[0.229, 0.224, 0.225]),
	1528		])
	1529
	1530		loader = torch.utils.data.DataLoader(PredictListDataset(
	1531		data_list,
	1532		shuffle=False,
	1533		transform=transformer
	1534		))
	1535
	1536		return loader

File debug/perfomance_test_on_shb.py changed (mode: 100644) (index 1de37c9..3233d29)
...	...	import time
24	24	import sys	import sys
25	25
26	26	if __name__ == "__main__":	if __name__ == "__main__":
	27		"""
	28		go for speed
	29		"""
27	30	torch.set_num_threads(2) # 4 thread	torch.set_num_threads(2) # 4 thread
28	31
29	32	device = torch.device("cuda" if torch.cuda.is_available() else "cpu")	device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

File demo_app/__init__.py copied from file debug/__init__.py (similarity 100%)

File demo_app/predict_only.py added (mode: 100644) (index 0000000..6d18a33)
	1	import os
	2	from data_flow import get_predict_dataloader
	3	if __name__ == "__main__":
	4	"""
	5	predict all in folder
	6	output into another folder
	7	output density map and count in csv
	8	"""
	9	INPUT_FOLDER = "/data/ShanghaiTech/part_B/test_data/images/"
	10	OUTPUT_FOLDER = "/data/apps/tmp"
	11	input_list = [os.path.join(INPUT_FOLDER, dir) for dir in os.listdir(INPUT_FOLDER)]
	12	loader = get_predict_dataloader(input_list)
	13	for img, info in loader:
	14	print(img.shape)
	15	print(info)

Hints:
Before first commit, do not forget to setup your git environment:

git config --global user.name "your_name_here"
git config --global user.email "your@email_here"

Clone this repository using HTTP(S):

git clone https://rocketgit.com/user/hahattpro/crowd_counting_framework

Clone this repository using ssh (do not forget to upload a key first):

git clone ssh://rocketgit@ssh.rocketgit.com/user/hahattpro/crowd_counting_framework

Clone this repository using git:

git clone git://git.rocketgit.com/user/hahattpro/crowd_counting_framework

You are allowed to anonymously push to this repository.
This means that your pushed commits will automatically be transformed into a merge request:

... clone the repository ...
... make some changes and some commits ...
git push origin main