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)
new ccnn	44a669c1f918be9d74313f29a5dbbc876c29f2fc	Thai Thien	2020-03-09 17:16:49
fix script	aa331331b12e5b454d372a550524b30a4bebe706	Thai Thien	2020-03-07 18:32:06
try reproduct ccnn with keepfull and lr 1e-5	814c520cbd1bb2d7fd50d2a8d3579d43da79fe60	Thai Thien	2020-03-07 18:30:42
my simple v4 with addition deform cnn at the end	5392aaf6c14fdd910f52096dbb921bed7470c4f7	Thai Thien	2020-03-07 18:15:22
fix the scheduler	77e6737a040f5aa5745b8a8830f5bec12322b10f	Thai Thien	2020-03-07 17:46:02
t4 lr 1e-5	acd41ed30c95f63e01a05a6d9929410637852d9e	Thai Thien	2020-03-06 19:41:49
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

Commit 44a669c1f918be9d74313f29a5dbbc876c29f2fc - new ccnn
Author: Thai Thien
Author date (UTC): 2020-03-09 17:16
Committer name: Thai Thien
Committer date (UTC): 2020-03-09 17:16
Parent(s): aa331331b12e5b454d372a550524b30a4bebe706
Signing key:
Tree: 49ae38e2f210bc15caeb5bea151c6698f017b2c3

File	Lines added	Lines deleted
models/__init__.py	1	1
models/compact_cnn.py	49	0
train_custom_compact_cnn.py	2	2

File models/__init__.py changed (mode: 100644) (index e02e0ec..a83ed52)
...	...	from .deform_conv_v2 import DeformConv2d
5	5	from .attn_can_adcrowdnet import AttnCanAdcrowdNet	from .attn_can_adcrowdnet import AttnCanAdcrowdNet
6	6	from .attn_can_adcrowdnet_freeze_vgg import AttnCanAdcrowdNetFreezeVgg	from .attn_can_adcrowdnet_freeze_vgg import AttnCanAdcrowdNetFreezeVgg
7	7	from .attn_can_adcrowdnet_simple import AttnCanAdcrowdNetSimpleV1, AttnCanAdcrowdNetSimpleV2, AttnCanAdcrowdNetSimpleV3, AttnCanAdcrowdNetSimpleV4	from .attn_can_adcrowdnet_simple import AttnCanAdcrowdNetSimpleV1, AttnCanAdcrowdNetSimpleV2, AttnCanAdcrowdNetSimpleV3, AttnCanAdcrowdNetSimpleV4
8		from .compact_cnn import CompactCNN, CompactDilatedCNN
	8		from .compact_cnn import CompactCNN, CompactDilatedCNN, DefDilatedCCNN

File models/compact_cnn.py changed (mode: 100644) (index 78d918a..3521e86)
1	1	import torch.nn as nn	import torch.nn as nn
2	2	import torch	import torch
3	3	from torchvision import models	from torchvision import models
	4		from .deform_conv_v2 import DeformConv2d
4	5	import collections	import collections
5	6	import torch.nn.functional as F	import torch.nn.functional as F
6	7

...	...	class CompactCNN(nn.Module):
15	16	self.red_cnn = nn.Conv2d(3, 10, 9, padding=4)	self.red_cnn = nn.Conv2d(3, 10, 9, padding=4)
16	17	self.green_cnn = nn.Conv2d(3, 14, 7, padding=3)	self.green_cnn = nn.Conv2d(3, 14, 7, padding=3)
17	18	self.blue_cnn = nn.Conv2d(3, 16, 5, padding=2)	self.blue_cnn = nn.Conv2d(3, 16, 5, padding=2)
	19
	20		self.red_cnn = DeformConv2d(3, 10, 9, padding=4)
	21		self.green_cnn = DeformConv2d(3, 14, 7, padding=3)
	22		self.blue_cnn = DeformConv2d(3, 16, 5, padding=2)
	23
18	24	self.max_pooling = nn.MaxPool2d(kernel_size=2, stride=2)	self.max_pooling = nn.MaxPool2d(kernel_size=2, stride=2)
19	25
20	26	self.c1 = nn.Conv2d(40, 60, 3, padding=1)	self.c1 = nn.Conv2d(40, 60, 3, padding=1)

...	...	class CompactDilatedCNN(nn.Module):
78	84	x = self.output(x)	x = self.output(x)
79	85	return x	return x
80	86
	87
	88		class DefDilatedCCNN(nn.Module):
	89		"""
	90
	91		"""
	92		def __init__(self, load_weights=False):
	93		super(DefDilatedCCNN, self).__init__()
	94
	95		self.red_cnn = DeformConv2d(3, 10, 9, padding=4)
	96		self.green_cnn = DeformConv2d(3, 14, 7, padding=3)
	97		self.blue_cnn = DeformConv2d(3, 16, 5, padding=2)
	98
	99		self.max_pooling = nn.MaxPool2d(kernel_size=2, stride=2)
	100
	101		self.c1 = nn.Conv2d(40, 60, 3, dilation=2, padding=2, bias=False)
	102		self.bn1 = nn.BatchNorm2d(60)
	103		self.c2 = nn.Conv2d(60, 40, 3, dilation=2, padding=2, bias=False)
	104		self.bn2 = nn.BatchNorm2d(40)
	105		self.c3 = nn.Conv2d(40, 20, 3, dilation=2, padding=2, bias=False)
	106		self.bn3 = nn.BatchNorm2d(20)
	107		self.c4 = nn.Conv2d(20, 10, 3, dilation=2, padding=2, bias=False)
	108		self.bn4 = nn.BatchNorm2d(10)
	109		self.output = nn.Conv2d(10, 1, 1)
	110
	111		def forward(self,x):
	112		x_red = self.max_pooling(F.relu(self.red_cnn(x), inplace=True))
	113		x_green = self.max_pooling(F.relu(self.green_cnn(x), inplace=True))
	114		x_blue = self.max_pooling(F.relu(self.blue_cnn(x), inplace=True))
	115
	116		x = torch.cat((x_red, x_green, x_blue), 1)
	117		x = F.relu(self.bn1(self.c1(x)), inplace=True)
	118
	119		x = F.relu(self.bn2(self.c2(x)), inplace=True)
	120		x = self.max_pooling(x)
	121
	122		x = F.relu(self.bn3(self.c3(x)), inplace=True)
	123		x = self.max_pooling(x)
	124
	125		x = F.relu(self.bn4(self.c4(x)), inplace=True)
	126
	127		x = self.output(x)
	128		return x
	129

File train_custom_compact_cnn.py changed (mode: 100644) (index 7c5d235..f28d251)
...	...	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 CompactDilatedCNN
	12		from models import DefDilatedCCNN
13	13	import os	import os
14	14
15	15	if __name__ == "__main__":	if __name__ == "__main__":

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

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