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)
we add head experiment	82a25e6b89945609486cbafc433eaea20dcdee39	Thai Thien	2020-04-27 16:23:50
CompactCNNV3	7de3766d085ebdbdf82b024eb517568dd82d8d6d	Thai Thien	2020-04-27 16:23:20
no_norm	da3c84dca19b0d281082679d88af3b9d27165bfe	Thai Thien	2020-04-25 17:32:45
M4_t3_sha_c_shb	d0d61ff74ed23f595d05d6a813c0a93239f61438	Thai Thien	2020-04-25 17:17:56
training script	624ecec7b12641f734e12ee2ebb6158c7c89683a	Thai Thien	2020-04-25 17:08:25
clean up trash	05fa10a45e7c4f9d0ba6b80e578a0f934a86121e	Thai Thien	2020-04-25 17:08:04
increase epoch sha	96e0315a34286751258902b1e954ea5e43145ee1	Thai Thien	2020-04-23 17:24:53
turn off debug	36a2603484395ed130dd7dcb69c98c7057adf3ec	Thai Thien	2020-04-23 17:21:51
chang proxy	ce5434adf6002e3c1e14eee8fb50023ba9f1da39	Thai Thien	2020-04-23 17:19:00
fix typo	4f5c034e3fcbd9d3dd66333c7e81db79053c210b	Thai Thien	2020-04-23 17:12:23
M4_t3	9ed8f35d94e43d0869d9ee07d98395ed4d4cd3fd	Thai Thien	2020-04-23 16:18:49
fix	b2235d1abdb20c3e1bfcc0d42870dad4b706babc	Thai Thien	2020-04-23 16:13:17
debug	6cda7a90f14c5768be5bfbb9d87a985cae845f4c	Thai Thien	2020-04-23 11:27:53
best score checkpoint , timer	bb26cec915aa04d68a0dd00911c273542f9b34b5	Thai Thien	2020-04-23 11:16:24
typo	2636cf5b78f062c89196c4f462afc4b72aa39798	Thai Thien	2020-04-19 12:10:51
M4	317900419b7ba25c679dd582a6de5fc00fc764ec	Thai Thien	2020-04-19 12:09:51
m4 t2	a02b9610e868f4ba5e64496dc0c861a269f4cb9f	Thai Thien	2020-04-17 16:01:39
fix url	358f164d558dab393f65c0829d8d9c37b1437ff3	Thai Thien	2020-04-16 14:32:49
increase epoch	03be68a9e02df1ffa245394ea3096990e8f9d44b	Thai Thien	2020-04-16 14:30:15
add load model	044a398d62add2e854b79b0b3c48c961a4a20bb0	Thai Thien	2020-04-16 14:27:43

Commit 82a25e6b89945609486cbafc433eaea20dcdee39 - we add head experiment

We replace 5x5 7x7 9x9 with 3x3, no batchnorm yet, keep tail, no dilated

Author: Thai Thien
Author date (UTC): 2020-04-27 16:23
Committer name: Thai Thien
Committer date (UTC): 2020-04-27 16:23
Parent(s): 7de3766d085ebdbdf82b024eb517568dd82d8d6d
Signing key:
Tree: 8b8a2c67344e2cbcdb396b60e1e27e98d1524ccd

File	Lines added	Lines deleted
models/meow_experiment/ccnn_head.py	75	0

File models/meow_experiment/ccnn_head.py added (mode: 100644) (index 0000000..1d2e772)
	1	import torch.nn as nn
	2	import torch
	3	import collections
	4	import torch.nn.functional as F
	5
	6
	7	class H1(nn.Module):
	8	"""
	9	A REAL-TIME DEEP NETWORK FOR CROWD COUNTING
	10	https://arxiv.org/pdf/2002.06515.pdf
	11	the improve version
	12
	13	we change 5x5 7x7 9x9 with 3x3
	14	Keep the tail
	15	"""
	16	def __init__(self, load_weights=False):
	17	super(H1, self).__init__()
	18	self.model_note = "We replace 5x5 7x7 9x9 with 3x3, no batchnorm yet, keep tail, no dilated"
	19	# self.red_cnn = nn.Conv2d(3, 10, 9, padding=4)
	20	# self.green_cnn = nn.Conv2d(3, 14, 7, padding=3)
	21	# self.blue_cnn = nn.Conv2d(3, 16, 5, padding=2)
	22
	23	# ideal from crowd counting using DMCNN
	24	self.front_cnn_1 = nn.Conv2d(3, 10, 3, padding=1)
	25	self.front_cnn_2 = nn.Conv2d(10, 20, 3, padding=1)
	26	self.front_cnn_3 = nn.Conv2d(20, 20, 3, padding=1)
	27	self.front_cnn_4 = nn.Conv2d(20, 20, 3, padding=1)
	28
	29	self.c0 = nn.Conv2d(60, 40, 3, padding=1)
	30	self.max_pooling = nn.MaxPool2d(kernel_size=2, stride=2)
	31
	32	self.c1 = nn.Conv2d(40, 60, 3, padding=1)
	33
	34	# ideal from CSRNet
	35	self.c2 = nn.Conv2d(60, 40, 3, padding=1)
	36	self.c3 = nn.Conv2d(40, 20, 3, padding=1)
	37	self.c4 = nn.Conv2d(20, 10, 3, padding=1)
	38	self.output = nn.Conv2d(10, 1, 1)
	39
	40	def forward(self,x):
	41	#x_red = self.max_pooling(F.relu(self.red_cnn(x), inplace=True))
	42	#x_green = self.max_pooling(F.relu(self.green_cnn(x), inplace=True))
	43	#x_blue = self.max_pooling(F.relu(self.blue_cnn(x), inplace=True))
	44
	45	x_red = F.relu(self.front_cnn_1(x), inplace=True)
	46	x_red = F.relu(self.front_cnn_2(x_red), inplace=True)
	47	x_red = F.relu(self.front_cnn_3(x_red), inplace=True)
	48	x_red = F.relu(self.front_cnn_4(x_red), inplace=True)
	49	x_red = self.max_pooling(x_red)
	50
	51	x_green = F.relu(self.front_cnn_1(x), inplace=True)
	52	x_green = F.relu(self.front_cnn_2(x_green), inplace=True)
	53	x_green = F.relu(self.front_cnn_3(x_green), inplace=True)
	54	x_green = self.max_pooling(x_green)
	55
	56	x_blue = F.relu(self.front_cnn_1(x), inplace=True)
	57	x_blue = F.relu(self.front_cnn_2(x_blue), inplace=True)
	58	x_blue = self.max_pooling(x_blue)
	59
	60	# x = self.max_pooling(x)
	61	x = torch.cat((x_red, x_green, x_blue), 1)
	62	x = F.relu(self.c0(x), inplace=True)
	63
	64	x = F.relu(self.c1(x), inplace=True)
	65
	66	x = F.relu(self.c2(x), inplace=True)
	67	x = self.max_pooling(x)
	68
	69	x = F.relu(self.c3(x), inplace=True)
	70	x = self.max_pooling(x)
	71
	72	x = F.relu(self.c4(x), inplace=True)
	73
	74	x = self.output(x)
	75	return x

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