Skip to content

Commit

Permalink
up training bisenet
Browse files Browse the repository at this point in the history
  • Loading branch information
@Vitabile
Vitabile committed Jan 7, 2024
1 parent 4c6798e commit 703ada9
Show file tree
Hide file tree
Showing 4 changed files with 246 additions and 0 deletions.
63 changes: 63 additions & 0 deletions save/bisenet_training1/automated_log.txt
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,63 @@
Epoch Train-loss Test-loss Train-IoU Test-IoU learningRate
1 3.9792 3.9001 0.0000 0.3599 0.00050000
2 3.5550 3.3912 0.0000 0.4236 0.00049700
3 3.4146 3.3582 0.0000 0.4412 0.00049400
4 3.3489 3.5177 0.0000 0.4532 0.00049099
5 3.2692 3.6933 0.0000 0.4839 0.00048798
6 3.2231 3.5844 0.0000 0.4533 0.00048497
7 3.1911 4.1798 0.0000 0.4076 0.00048196
8 3.1515 3.2479 0.0000 0.5060 0.00047895
9 3.1084 3.5135 0.0000 0.5562 0.00047593
10 3.1014 3.5033 0.0000 0.5431 0.00047292
11 3.0681 3.7079 0.0000 0.5293 0.00046990
12 3.0532 3.2032 0.0000 0.5600 0.00046688
13 3.0484 3.3251 0.0000 0.5934 0.00046385
14 3.0155 3.8794 0.0000 0.4733 0.00046083
15 3.0064 3.2069 0.0000 0.5381 0.00045780
16 2.9752 3.5603 0.0000 0.5861 0.00045477
17 2.9469 3.5183 0.0000 0.5954 0.00045173
18 2.9506 3.4961 0.0000 0.5658 0.00044870
19 2.9242 3.4800 0.0000 0.5621 0.00044566
20 2.9282 3.8263 0.0000 0.5277 0.00044262
21 2.9019 3.6284 0.0000 0.5649 0.00043958
22 2.8859 3.4999 0.0000 0.5970 0.00043653
23 2.8831 3.8472 0.0000 0.5607 0.00043349
24 2.8798 3.8634 0.0000 0.5826 0.00043044
25 2.8788 3.8610 0.0000 0.5682 0.00042739
26 2.8386 3.7606 0.0000 0.5850 0.00042433
27 2.8220 3.8189 0.0000 0.6167 0.00042128
28 2.8052 3.5966 0.0000 0.6143 0.00041822
29 2.7949 3.8360 0.0000 0.5768 0.00041516
30 2.8160 3.7362 0.0000 0.6112 0.00041209
31 2.7986 3.7714 0.0000 0.5650 0.00040903
32 2.7979 3.8223 0.0000 0.6013 0.00040596
33 2.7663 4.0719 0.0000 0.5800 0.00040289
34 2.7599 4.1707 0.0000 0.5984 0.00039981
35 2.7504 3.9280 0.0000 0.5985 0.00039673
36 2.7713 3.8951 0.0000 0.5953 0.00039366
37 2.7262 4.0748 0.0000 0.6329 0.00039057
38 2.7137 3.9763 0.0000 0.6288 0.00038749
39 2.7054 4.0973 0.0000 0.6214 0.00038440
40 2.7175 4.2663 0.0000 0.5983 0.00038131
41 2.6948 3.9869 0.0000 0.6177 0.00037822
42 2.6946 4.1193 0.0000 0.6079 0.00037512
43 2.6772 4.1540 0.0000 0.6095 0.00037202
44 2.6852 4.1108 0.0000 0.6114 0.00036892
45 2.7099 4.7217 0.0000 0.5490 0.00036582
46 2.6982 4.1744 0.0000 0.6263 0.00036271
47 2.6520 4.1139 0.0000 0.6226 0.00035960
48 2.6393 4.5040 0.0000 0.6379 0.00035649
49 2.6345 4.5977 0.0000 0.5833 0.00035337
50 2.6514 4.3122 0.0000 0.6229 0.00035025
51 2.6691 4.2598 0.0000 0.6107 0.00034713
52 2.6328 4.2943 0.0000 0.6095 0.00034400
53 2.6049 4.5418 0.0000 0.6112 0.00034087
54 2.6121 4.2598 0.0000 0.6026 0.00033774
55 2.6286 4.3250 0.0000 0.6064 0.00033460
56 2.5921 4.5597 0.0000 0.6224 0.00033147
57 2.5757 4.5647 0.0000 0.6242 0.00032832
58 2.5736 4.5556 0.0000 0.6137 0.00032518
59 2.6799 3.9880 0.0000 0.5662 0.00032203
60 2.6341 4.2541 0.0000 0.6363 0.00031888
61 2.5860 4.4965 0.0000 0.6342 0.00031572
62 2.5570 4.7645 0.0000 0.6327 0.00031256
1 change: 1 addition & 0 deletions save/bisenet_training1/best.txt
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
Best epoch is 48, with Val-IoU= 0.6379
181 changes: 181 additions & 0 deletions save/bisenet_training1/model.txt
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,181 @@
DataParallel(
(module): Net(
(cp): ContextPath(
(resnet): Resnet18(
(conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
(layer1): Sequential(
(0): BasicBlock(
(conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(1): BasicBlock(
(conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
)
(layer2): Sequential(
(0): BasicBlock(
(conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(downsample): Sequential(
(0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): BasicBlock(
(conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
)
(layer3): Sequential(
(0): BasicBlock(
(conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(downsample): Sequential(
(0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): BasicBlock(
(conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
)
(layer4): Sequential(
(0): BasicBlock(
(conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(downsample): Sequential(
(0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): BasicBlock(
(conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
)
)
(arm16): AttentionRefinementModule(
(conv): ConvBNReLU(
(conv): Conv2d(256, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv_atten): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn_atten): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
(arm32): AttentionRefinementModule(
(conv): ConvBNReLU(
(conv): Conv2d(512, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv_atten): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn_atten): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
(conv_head32): ConvBNReLU(
(conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv_head16): ConvBNReLU(
(conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv_avg): ConvBNReLU(
(conv): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(up32): Upsample(scale_factor=2.0, mode='nearest')
(up16): Upsample(scale_factor=2.0, mode='nearest')
)
(sp): SpatialPath(
(conv1): ConvBNReLU(
(conv): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv2): ConvBNReLU(
(conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv3): ConvBNReLU(
(conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv_out): ConvBNReLU(
(conv): Conv2d(64, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
)
(ffm): FeatureFusionModule(
(convblk): ConvBNReLU(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
(conv_out): BiSeNetOutput(
(conv): ConvBNReLU(
(conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv_out): Conv2d(256, 20, kernel_size=(1, 1), stride=(1, 1))
(up): Upsample(scale_factor=8.0, mode='bilinear')
)
(conv_out16): BiSeNetOutput(
(conv): ConvBNReLU(
(conv): Conv2d(128, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv_out): Conv2d(64, 20, kernel_size=(1, 1), stride=(1, 1))
(up): Upsample(scale_factor=8.0, mode='bilinear')
)
(conv_out32): BiSeNetOutput(
(conv): ConvBNReLU(
(conv): Conv2d(128, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
)
(conv_out): Conv2d(64, 20, kernel_size=(1, 1), stride=(1, 1))
(up): Upsample(scale_factor=16.0, mode='bilinear')
)
)
)
1 change: 1 addition & 0 deletions save/bisenet_training1/opts.txt
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
Namespace(cuda=True, model='bisenetv1', state=None, port=8097, datadir='C:/Users/David/Documents/Polito/Github Repo/Real-Time-Anomaly-Segmentation-for-Road-Scenes//cityscapes/', height=512, num_epochs=150, num_workers=4, batch_size=8, steps_loss=50, steps_plot=50, epochs_save=0, savedir='bisenet_training1', decoder=False, pretrainedEncoder=None, visualize=True, iouTrain=False, iouVal=True, resume=False, erfnet=False)

0 comments on commit 703ada9

Please sign in to comment.