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)
abs clamp	524c6f7196908527367ddcbfc340107e4092cc6e	Thai Thien	2020-12-06 11:16:12
use abs instead of ssim	92226d63c9732e66bf50f0b78e61f915ad2bce77	Thai Thien	2020-12-06 10:56:56
claim both y and y_pred	d1f6764a691fc2f2f1adc63ae65d63a8347d8a45	Thai Thien	2020-12-06 10:52:14
clamp 0.0	5569f7dad3c9e792e65636bb29b66a84ff250282	Thai Thien	2020-12-06 10:46:57
dataloader target crop	e41c17147de7c05f213329009c7662fbb2f1dc91	Thai Thien	2020-12-06 10:43:12
clamp min = 0	5245898e5282aa4cdc4b6539e68cb36a1c2d0c1f	Thai Thien	2020-12-06 10:40:47
fix metric	9739b99b836657377eb9dce05fafeef4a90546e9	Thai Thien	2020-12-06 10:38:37
eval density	d02232a419cd22bc79e0325c1c6791a7e5fc15b0	Thai Thien	2020-12-06 10:37:22
.	926498e13908770b072aeca0ccecbf5d3a64808e	Thai Thien	2020-12-06 10:27:13
if self.dataset_name == "shanghaitech_non_overlap_test_with_densitygt" and not self.train:	ca2bf6f9372ccdd5b880f2dd7be2023840afdd1d	Thai Thien	2020-12-06 10:24:21
yeah forgot to change ds name	bfdf3cfe27be5f51de300e231b6b79f903a71624	Thai Thien	2020-12-06 10:17:37
eval_only mode should not save	bf03ec80b43e070501c043f7428d55d8409b960c	Thai Thien	2020-12-06 10:14:22
now we output density map for testing	fb1f9f1c8abeb2311f0637db5f7b4c4ca4b43810	Thai Thien	2020-12-06 10:12:01
predict	8a312e6bcbe7a8b726cf1a01497db09bc21fea51	Thai Thien	2020-12-06 09:57:58
go ssim and psnr	d9b0d9c0417835c51ea59c66d85b8821480d0357	Thai Thien	2020-12-06 09:36:21
t4	624c9948c2099f16bd66f30a0c18093367b8a113	Thai Thien	2020-12-06 03:59:20
adamw1_ccnnv7_t3_bike	944e85901a2c22fd09f771cd1dd295f5a0ead80b	Thai Thien	2020-12-06 03:51:57
train ccnn baseline on mybike ds	72d6a45b37523136c36ef208e7c9ab0893bce3da	Thai Thien	2020-12-06 03:25:17
my collate on my bike	748f11c4b2ddbd37e5dda3b9472558fa49689c1f	Thai Thien	2020-12-05 19:06:48
pump up the epoch	5b4e119e1f1eaf56d51e0dba9dd89ecc2a38ad5b	Thai Thien	2020-12-05 19:03:41

Commit 524c6f7196908527367ddcbfc340107e4092cc6e - abs clamp
Author: Thai Thien
Author date (UTC): 2020-12-06 11:16
Committer name: Thai Thien
Committer date (UTC): 2020-12-06 11:16
Parent(s): 92226d63c9732e66bf50f0b78e61f915ad2bce77
Signer:
Signing key:
Signing status: N
Tree: 7c314cdf1e0581eccbb72774ffddcd08908910c8

File	Lines added	Lines deleted
crowd_counting_error_metrics.py	64	2
experiment_main.py	18	7

File crowd_counting_error_metrics.py changed (mode: 100644) (index ea8ceba..f8d2d31)
...	...	class CrowdCountingMeanSquaredErrorWithCount(Metric):
112	112	import piq	import piq
113	113
114	114
115		class CrowdCountingMeanSSIM(Metric):
	115		class CrowdCountingMeanSSIMabs(Metric):
116	116	"""	"""
117	117	Calculates ssim	Calculates ssim
118	118	require package https://github.com/photosynthesis-team/piq	require package https://github.com/photosynthesis-team/piq

...	...	class CrowdCountingMeanSSIM(Metric):
144	144	return self._sum / self._num_examples	return self._sum / self._num_examples
145	145
146	146
147		class CrowdCountingMeanPSNR(Metric):
	147		class CrowdCountingMeanPSNRabs(Metric):
148	148	"""	"""
149	149	Calculates ssim	Calculates ssim
150	150	require package https://github.com/photosynthesis-team/piq	require package https://github.com/photosynthesis-team/piq

...	...	class CrowdCountingMeanPSNR(Metric):
176	176	raise NotComputableError('CrowdCountingMeanPSNR must have at least one example before it can be computed.')	raise NotComputableError('CrowdCountingMeanPSNR must have at least one example before it can be computed.')
177	177	return self._sum / self._num_examples	return self._sum / self._num_examples
178	178
	179		#################3
	180
	181
	182		class CrowdCountingMeanSSIMclamp(Metric):
	183		"""
	184		Calculates ssim
	185		require package https://github.com/photosynthesis-team/piq
	186		- `update` must receive output of the form `(y_pred, y)`.
	187		"""
	188		def reset(self):
	189		self._sum = 0.0
	190		self._num_examples = 0
	191
	192		def update(self, output):
	193		y_pred = output[0]
	194		y = output[1]
	195		y_pred = torch.clamp_min(y_pred, min=0.0)
	196		y = torch.clamp_min(y, min=0.0)
	197
	198
	199		ssim_metric = piq.ssim(y, y_pred)
	200
	201		self._sum += ssim_metric.item() * y.shape[0]
	202		# we multiply because ssim calculate mean of each image in batch
	203		# we multiply so we will divide correctly
	204
	205		self._num_examples += y.shape[0]
	206
	207		def compute(self):
	208		if self._num_examples == 0:
	209		raise NotComputableError('CrowdCountingMeanSSIM must have at least one example before it can be computed.')
	210		return self._sum / self._num_examples
	211
	212
	213		class CrowdCountingMeanPSNRclamp(Metric):
	214		"""
	215		Calculates ssim
	216		require package https://github.com/photosynthesis-team/piq
	217		- `update` must receive output of the form `(y_pred, y)`.
	218		"""
	219		def reset(self):
	220		self._sum = 0.0
	221		self._num_examples = 0
	222
	223		def update(self, output):
	224		y_pred = output[0]
	225		y_pred = torch.clamp_min(y_pred, min=0.0)
	226		y = output[1]
	227		y = torch.clamp_min(y, min=0.0)
	228
	229		psnr_metric = piq.psnr(y, y_pred)
	230
	231		self._sum += psnr_metric.item() * y.shape[0]
	232		# we multiply because ssim calculate mean of each image in batch
	233		# we multiply so we will divide correctly
	234
	235		self._num_examples += y.shape[0]
	236
	237		def compute(self):
	238		if self._num_examples == 0:
	239		raise NotComputableError('CrowdCountingMeanPSNR must have at least one example before it can be computed.')
	240		return self._sum / self._num_examples

File experiment_main.py changed (mode: 100644) (index 5210431..10aac00)
...	...	from data_flow import get_dataloader, create_image_list
5	5	from ignite.engine import Events, create_supervised_trainer, create_supervised_evaluator	from ignite.engine import Events, create_supervised_trainer, create_supervised_evaluator
6	6	from ignite.metrics import Loss	from ignite.metrics import Loss
7	7	from ignite.handlers import Checkpoint, DiskSaver, Timer	from ignite.handlers import Checkpoint, DiskSaver, Timer
8		from crowd_counting_error_metrics import CrowdCountingMeanAbsoluteError, CrowdCountingMeanSquaredError, CrowdCountingMeanAbsoluteErrorWithCount, CrowdCountingMeanSquaredErrorWithCount, CrowdCountingMeanSSIM, CrowdCountingMeanPSNR
	8		from crowd_counting_error_metrics import CrowdCountingMeanAbsoluteError, CrowdCountingMeanSquaredError,\
	9		CrowdCountingMeanAbsoluteErrorWithCount, CrowdCountingMeanSquaredErrorWithCount,\
	10		CrowdCountingMeanSSIMabs, CrowdCountingMeanPSNRabs, \
	11		CrowdCountingMeanSSIMclamp, CrowdCountingMeanPSNRclamp
	12
9	13	from visualize_util import get_readable_time	from visualize_util import get_readable_time
10	14	from mse_l1_loss import MSEL1Loss, MSE4L1Loss	from mse_l1_loss import MSEL1Loss, MSE4L1Loss
11	15	import torch	import torch

...	...	if __name__ == "__main__":
228	232	if args.eval_density:	if args.eval_density:
229	233	evaluator_test = create_supervised_evaluator(model,	evaluator_test = create_supervised_evaluator(model,
230	234	metrics={	metrics={
231		'ssim': CrowdCountingMeanSSIM(),
232		'psnr': CrowdCountingMeanPSNR(),
	235		'ssimabs': CrowdCountingMeanSSIMabs(),
	236		'psnrabs': CrowdCountingMeanPSNRabs(),
	237		'ssimclamp': CrowdCountingMeanSSIMclamp(),
	238		'psnrclamp': CrowdCountingMeanPSNRclamp(),
233	239	}, device=device)	}, device=device)
234	240	else:	else:
235	241	evaluator_test = create_supervised_evaluator(model,	evaluator_test = create_supervised_evaluator(model,

...	...	if __name__ == "__main__":
371	377	timestamp = get_readable_time()	timestamp = get_readable_time()
372	378
373	379	if args.eval_density:	if args.eval_density:
374		print(timestamp + " Test set Results - Avg ssim: {:.2f} Avg psnr: {:.2f} Avg loss: {:.2f}"
375		.format(test_metrics['ssim'], test_metrics['psnr'], 0))
376		experiment.log_metric("test_ssim", test_metrics['ssim'])
377		experiment.log_metric("test_psnr", test_metrics['psnr'])
	380		print(timestamp + " Test set Results ABS - Avg ssim: {:.2f} Avg psnr: {:.2f} Avg loss: {:.2f}"
	381		.format(test_metrics['ssimabs'], test_metrics['psnrabs'], 0))
	382		experiment.log_metric("test_ssim abs", test_metrics['ssimabs'])
	383		experiment.log_metric("test_psnr abs", test_metrics['psnrabs'])
	384
	385		print(timestamp + " Test set Results CLAMP - Avg ssim: {:.2f} Avg psnr: {:.2f} Avg loss: {:.2f}"
	386		.format(test_metrics['ssimclamp'], test_metrics['psnrclamp'], 0))
	387		experiment.log_metric("test_ssim clamp", test_metrics['ssimclamp'])
	388		experiment.log_metric("test_psnr clamp", test_metrics['psnrclamp'])
378	389	else:	else:
379	390	print(timestamp + " Test set Results - Avg mae: {:.2f} Avg mse: {:.2f} Avg loss: {:.2f}"	print(timestamp + " Test set Results - Avg mae: {:.2f} Avg mse: {:.2f} Avg loss: {:.2f}"
380	391	.format( test_metrics['mae'], test_metrics['mse'], 0))	.format( test_metrics['mae'], test_metrics['mse'], 0))

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