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)
no more lr scheduler	7289adb41de7807258eb8c29e6108fa65f59525a	Thai Thien	2020-03-06 19:35:49
reduce learning rate	bc8241e5b88b91c18bb7999a8d5d12fc79a5e3f7	Thai Thien	2020-03-06 19:28:27
dilated ccnn	5c5d92bdc0a288dd5d4ec5f1367d8cb928175bbe	Thai Thien	2020-03-06 19:04:01
done	9f05e093ec7c10284a4aedf0738f9e61d5ac6fb6	Thai Thien	2020-03-06 18:02:34
with lr scheduler	466c364b60ed22c77319b14ccc9a201614b908bf	Thai Thien	2020-03-04 17:57:49
train with learning rate scheduler	fcd5a3c8da2dd6763e0d40742edf47b49c95fcfb	Thai Thien	2020-03-04 17:55:11
ccnn no padding at output layer	57563fc07f656c63f807de4d80712ff11345109d	Thai Thien	2020-03-04 17:36:43
fix dimension of ccnn	f4439d9a78273ab3ba450f31a528509816b4352f	Thai Thien	2020-03-04 17:32:48
ready to train	dbe0d6c3271dbb22490f0877fa31ba9cd7852b99	Thai Thien	2020-03-04 15:55:05
done implement c-cnn	2deecef953baf1e07ce5cf5477d208bc7ffa34cf	Thai Thien	2020-03-03 17:25:03
fix script	0e9d372b9ad60b32939f1e558b2a59fc7d518fa2	Thai Thien	2020-03-02 16:23:55
simple v3 to 91 epoch	539fdd03c3e3497fd22b7db2aaa14f067cbf6f8d	Thai Thien	2020-03-02 16:09:43
we train on all training data and validate on test data	9407ef8d5b7c47c53d6f98dcb3c20208aad1d7a9	Thai Thien	2020-03-01 15:36:46
load and continue train v3	12421fb7330e5c9d2eed4f6e574dfe69bdfddefc	Thai Thien	2020-03-01 14:50:01
add env file	b1ed02088b01af42efc8d6963b3699e0a5c31c01	Thai Thien	2020-03-01 11:10:56
sanity check dataloader	034daa8bef69daff92891cc42b988a6c77b010f9	Thai Thien	2020-03-01 10:24:38
print train loader len	eef3995f63a631e0ec5d92e31f5d7db27fd04401	Thai Thien	2020-03-01 05:17:55
simple v3 t1	9ee69fce793709c3dcc692de73ddb9d01c54670e	Thai Thien	2020-03-01 05:08:40
try fix load model not training	a9d6dc79a460f15bd34bc7c6a4e1359209f59f27	Thai Thien	2020-03-01 05:04:48
continue training for attn can adcrowdnet simple	437fe77a4f50b6c6098a5251dbb42be74d8cdfc4	Thai Thien	2020-03-01 04:40:18

Commit 7289adb41de7807258eb8c29e6108fa65f59525a - no more lr scheduler
Author: Thai Thien
Author date (UTC): 2020-03-06 19:35
Committer name: Thai Thien
Committer date (UTC): 2020-03-06 19:35
Parent(s): bc8241e5b88b91c18bb7999a8d5d12fc79a5e3f7
Signer:
Signing key:
Signing status: N
Tree: f70a67b05ae61c792db94e6e726a85d046f35b1d

File	Lines added	Lines deleted
train_custom_compact_cnn.py	3	3
train_script/CCNN_custom/dilated_ccnn_v1_t2.sh	0	16
train_script/CCNN_custom/dilated_ccnn_v1_t3.sh	25	0

File train_custom_compact_cnn.py copied from file train_compact_cnn.py (similarity 98%) (mode: 100644) (index 1876089..7c5d235)
...	...	from visualize_util import get_readable_time
9	9
10	10	import torch	import torch
11	11	from torch import nn	from torch import nn
12		from models import CompactCNN
	12		from models import CompactDilatedCNN
13	13	import os	import os
14	14
15
16	15	if __name__ == "__main__":	if __name__ == "__main__":
17	16	device = torch.device("cuda" if torch.cuda.is_available() else "cpu")	device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
18	17	print(device)	print(device)

...	...	if __name__ == "__main__":
40	39	print("len train_loader ", len(train_loader))	print("len train_loader ", len(train_loader))
41	40
42	41	# model	# model
43		model = CompactCNN()
	42		model = CompactDilatedCNN()
44	43	model = model.to(device)	model = model.to(device)
45	44
46	45	# loss function	# loss function

...	...	if __name__ == "__main__":
49	48	optimizer = torch.optim.Adam(model.parameters(), args.lr,	optimizer = torch.optim.Adam(model.parameters(), args.lr,
50	49	weight_decay=args.decay)	weight_decay=args.decay)
51	50
	51
52	52	trainer = create_supervised_trainer(model, optimizer, loss_fn, device=device)	trainer = create_supervised_trainer(model, optimizer, loss_fn, device=device)
53	53	evaluator = create_supervised_evaluator(model,	evaluator = create_supervised_evaluator(model,
54	54	metrics={	metrics={

File train_script/CCNN_custom/dilated_ccnn_v1_t2.sh deleted (index 532d60f..0000000)
1		#CUDA_VISIBLE_DEVICES=5 nohup python train_custom_compact_cnn_lrscheduler.py \
2		#--task_id dilated_ccnn_v1_t1 \
3		#--input /data/rnd/thient/thient_data/ShanghaiTech/part_A \
4		#--lr 1e-4 \
5		#--decay 5e-4 \
6		#--datasetname shanghaitech \
7		#--epochs 400 > logs/dilated_ccnn_v1_t1.log &
8
9
10		CUDA_VISIBLE_DEVICES=5 nohup python train_custom_compact_cnn_lrscheduler.py \
11		--task_id dilated_ccnn_v1_t2 \
12		--input /data/rnd/thient/thient_data/ShanghaiTech/part_A \
13		--lr 1e-5 \
14		--decay 5e-4 \
15		--datasetname shanghaitech \
16		--epochs 400 > logs/dilated_ccnn_v1_t2.log &

File train_script/CCNN_custom/dilated_ccnn_v1_t3.sh added (mode: 100644) (index 0000000..5036ae5)
	1	#CUDA_VISIBLE_DEVICES=5 nohup python train_custom_compact_cnn_lrscheduler.py \
	2	#--task_id dilated_ccnn_v1_t1 \
	3	#--input /data/rnd/thient/thient_data/ShanghaiTech/part_A \
	4	#--lr 1e-4 \
	5	#--decay 5e-4 \
	6	#--datasetname shanghaitech \
	7	#--epochs 400 > logs/dilated_ccnn_v1_t1.log &
	8
	9
	10	#CUDA_VISIBLE_DEVICES=5 nohup python train_custom_compact_cnn_lrscheduler.py \
	11	#--task_id dilated_ccnn_v1_t2 \
	12	#--input /data/rnd/thient/thient_data/ShanghaiTech/part_A \
	13	#--lr 1e-5 \
	14	#--decay 5e-4 \
	15	#--datasetname shanghaitech \
	16	#--epochs 400 > logs/dilated_ccnn_v1_t2.log &
	17
	18
	19	CUDA_VISIBLE_DEVICES=5 nohup python train_custom_compact_cnn.py \
	20	--task_id dilated_ccnn_v1_t3 \
	21	--input /data/rnd/thient/thient_data/ShanghaiTech/part_A \
	22	--lr 1e-6 \
	23	--decay 5e-5 \
	24	--datasetname shanghaitech \
	25	--epochs 400 > logs/dilated_ccnn_v1_t3.log &

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