Hugging Face's logo

Darayut
/
segformer-b2-detection

Transformers
Safetensors
segformer
Generated from Trainer
Model card Files
xet
 Community
Darayut commited on
Commit
9729fb4
·
verified ·
1 Parent(s): c6c61be

End of training

Browse files
Files changed (1) hide show
  1. README.md +44 -24
README.md CHANGED
@@ -14,8 +14,8 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model was trained from scratch on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 0.2188
18
- - Mean Iou: 0.3888
19
 
20
  ## Model description
21
 
@@ -34,40 +34,60 @@ More information needed
34
  ### Training hyperparameters
35
 
36
  The following hyperparameters were used during training:
37
- - learning_rate: 4e-05
38
  - train_batch_size: 8
39
  - eval_batch_size: 8
40
  - seed: 42
41
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
42
  - lr_scheduler_type: cosine
43
  - lr_scheduler_warmup_steps: 1000
44
- - training_steps: 20000
45
  - mixed_precision_training: Native AMP
46
 
47
  ### Training results
48
 
49
  | Training Loss | Epoch | Step | Validation Loss | Mean Iou |
50
  |:-------------:|:------:|:-----:|:---------------:|:--------:|
51
- | 0.1968 | 0.05 | 1000 | 0.2098 | 0.3513 |
52
- | 0.1468 | 0.1 | 2000 | 0.2712 | 0.3810 |
53
- | 0.1695 | 0.15 | 3000 | 0.2655 | 0.3701 |
54
- | 0.1684 | 0.2 | 4000 | 0.2394 | 0.3495 |
55
- | 0.1782 | 0.25 | 5000 | 0.2119 | 0.3878 |
56
- | 0.186 | 0.3 | 6000 | 0.2000 | 0.3791 |
57
- | 0.1576 | 0.35 | 7000 | 0.2072 | 0.3485 |
58
- | 0.1592 | 0.4 | 8000 | 0.2233 | 0.3833 |
59
- | 0.1641 | 1.0164 | 9000 | 0.2462 | 0.3698 |
60
- | 0.1567 | 1.0663 | 10000 | 0.2271 | 0.3579 |
61
- | 0.1484 | 1.1163 | 11000 | 0.2234 | 0.3895 |
62
- | 0.118 | 1.1663 | 12000 | 0.2352 | 0.3824 |
63
- | 0.1291 | 1.2164 | 13000 | 0.2003 | 0.3755 |
64
- | 0.1599 | 1.2664 | 14000 | 0.2186 | 0.3783 |
65
- | 0.1646 | 1.3163 | 15000 | 0.2267 | 0.3778 |
66
- | 0.1546 | 1.3663 | 16000 | 0.2221 | 0.3888 |
67
- | 0.1285 | 1.4163 | 17000 | 0.2162 | 0.3914 |
68
- | 0.1415 | 2.0327 | 18000 | 0.2174 | 0.3887 |
69
- | 0.173 | 2.0827 | 19000 | 0.2220 | 0.3889 |
70
- | 0.1364 | 2.1327 | 20000 | 0.2188 | 0.3888 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
71
 
72
 
73
  ### Framework versions
 
14
 
15
  This model was trained from scratch on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 0.2432
18
+ - Mean Iou: 0.3818
19
 
20
  ## Model description
21
 
 
34
  ### Training hyperparameters
35
 
36
  The following hyperparameters were used during training:
37
+ - learning_rate: 5e-05
38
  - train_batch_size: 8
39
  - eval_batch_size: 8
40
  - seed: 42
41
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
42
  - lr_scheduler_type: cosine
43
  - lr_scheduler_warmup_steps: 1000
44
+ - training_steps: 40000
45
  - mixed_precision_training: Native AMP
46
 
47
  ### Training results
48
 
49
  | Training Loss | Epoch | Step | Validation Loss | Mean Iou |
50
  |:-------------:|:------:|:-----:|:---------------:|:--------:|
51
+ | 0.1722 | 0.025 | 1000 | 0.2283 | 0.3572 |
52
+ | 0.1192 | 0.05 | 2000 | 0.2526 | 0.3822 |
53
+ | 0.1456 | 0.075 | 3000 | 0.2463 | 0.3827 |
54
+ | 0.154 | 0.1 | 4000 | 0.2447 | 0.3632 |
55
+ | 0.1688 | 0.125 | 5000 | 0.2146 | 0.3598 |
56
+ | 0.1725 | 0.15 | 6000 | 0.2148 | 0.3287 |
57
+ | 0.1522 | 0.175 | 7000 | 0.1987 | 0.3776 |
58
+ | 0.1454 | 0.2 | 8000 | 0.2618 | 0.3824 |
59
+ | 0.1631 | 1.0082 | 9000 | 0.3216 | 0.3516 |
60
+ | 0.1428 | 1.0332 | 10000 | 0.2662 | 0.3823 |
61
+ | 0.1439 | 1.0582 | 11000 | 0.2530 | 0.3502 |
62
+ | 0.1148 | 1.0832 | 12000 | 0.2817 | 0.3844 |
63
+ | 0.1299 | 1.1082 | 13000 | 0.2481 | 0.3618 |
64
+ | 0.155 | 1.1332 | 14000 | 0.2587 | 0.3718 |
65
+ | 0.166 | 1.1582 | 15000 | 0.2753 | 0.3708 |
66
+ | 0.1505 | 1.1832 | 16000 | 0.2401 | 0.4050 |
67
+ | 0.1293 | 1.2082 | 17000 | 0.2345 | 0.3551 |
68
+ | 0.1375 | 2.0164 | 18000 | 0.2423 | 0.3611 |
69
+ | 0.1721 | 2.0413 | 19000 | 0.2239 | 0.3950 |
70
+ | 0.1448 | 2.0663 | 20000 | 0.2273 | 0.3564 |
71
+ | 0.1067 | 2.0913 | 21000 | 0.2268 | 0.3851 |
72
+ | 0.1497 | 2.1164 | 22000 | 0.2340 | 0.3524 |
73
+ | 0.1436 | 2.1414 | 23000 | 0.2465 | 0.3718 |
74
+ | 0.1402 | 2.1663 | 24000 | 0.2669 | 0.3817 |
75
+ | 0.1224 | 2.1913 | 25000 | 0.2416 | 0.3895 |
76
+ | 0.1813 | 2.2163 | 26000 | 0.2375 | 0.3681 |
77
+ | 0.1361 | 3.0245 | 27000 | 0.2723 | 0.3833 |
78
+ | 0.1506 | 3.0495 | 28000 | 0.2838 | 0.3764 |
79
+ | 0.1367 | 3.0745 | 29000 | 0.2709 | 0.3473 |
80
+ | 0.1162 | 3.0995 | 30000 | 0.2685 | 0.3930 |
81
+ | 0.1803 | 3.1245 | 31000 | 0.2489 | 0.3886 |
82
+ | 0.0907 | 3.1495 | 32000 | 0.2487 | 0.3869 |
83
+ | 0.1158 | 3.1745 | 33000 | 0.2422 | 0.3885 |
84
+ | 0.1236 | 3.1995 | 34000 | 0.2326 | 0.3923 |
85
+ | 0.1412 | 4.0077 | 35000 | 0.2295 | 0.3885 |
86
+ | 0.1287 | 4.0327 | 36000 | 0.2405 | 0.3844 |
87
+ | 0.1525 | 4.0577 | 37000 | 0.2429 | 0.3857 |
88
+ | 0.133 | 4.0827 | 38000 | 0.2438 | 0.3829 |
89
+ | 0.146 | 4.1077 | 39000 | 0.2381 | 0.3840 |
90
+ | 0.1142 | 4.1327 | 40000 | 0.2432 | 0.3818 |
91
 
92
 
93
  ### Framework versions