Hugging Face's logo

ironchanchellor
/
segformer-b0_DsMetalDam_Augmented_Cropped

Transformers
PyTorch
segformer
Generated from Trainer
Inference Endpoints
Model card Files Community
1
ironchanchellor commited on
Commit
6b5da22
·
1 Parent(s): cec6cc1

segformer-b0_DsMetalDam_Augmented_Cropped

Browse files
Files changed (4) hide show
  1. README.md +99 -0
  2. config.json +84 -0
  3. pytorch_model.bin +3 -0
  4. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,99 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ base_model: nvidia/mit-b0
4
+ tags:
5
+ - generated_from_trainer
6
+ model-index:
7
+ - name: segformer-b0_DsMetalDam_Augmented_Cropped
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
+ should probably proofread and complete it, then remove this comment. -->
13
+
14
+ # segformer-b0_DsMetalDam_Augmented_Cropped
15
+
16
+ This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on an unknown dataset.
17
+ It achieves the following results on the evaluation set:
18
+ - Loss: 0.2486
19
+ - Mean Iou: 0.6867
20
+ - Mean Accuracy: 0.7623
21
+ - Overall Accuracy: 0.9106
22
+ - Accuracy Matrix: 0.8910
23
+ - Accuracy Austenite: 0.9442
24
+ - Accuracy Martensite/austenite: 0.8061
25
+ - Accuracy Precipitate: 0.2109
26
+ - Accuracy Defect: 0.9591
27
+ - Iou Matrix: 0.8022
28
+ - Iou Austenite: 0.8886
29
+ - Iou Martensite/austenite: 0.6946
30
+ - Iou Precipitate: 0.1697
31
+ - Iou Defect: 0.8786
32
+
33
+ ## Model description
34
+
35
+ More information needed
36
+
37
+ ## Intended uses & limitations
38
+
39
+ More information needed
40
+
41
+ ## Training and evaluation data
42
+
43
+ More information needed
44
+
45
+ ## Training procedure
46
+
47
+ ### Training hyperparameters
48
+
49
+ The following hyperparameters were used during training:
50
+ - learning_rate: 6e-05
51
+ - train_batch_size: 4
52
+ - eval_batch_size: 4
53
+ - seed: 42
54
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
55
+ - lr_scheduler_type: linear
56
+ - num_epochs: 30
57
+
58
+ ### Training results
59
+
60
+ | Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Matrix | Accuracy Austenite | Accuracy Martensite/austenite | Accuracy Precipitate | Accuracy Defect | Iou Matrix | Iou Austenite | Iou Martensite/austenite | Iou Precipitate | Iou Defect |
61
+ |:-------------:|:-----:|:-----:|:---------------:|:--------:|:-------------:|:----------------:|:---------------:|:------------------:|:-----------------------------:|:--------------------:|:---------------:|:----------:|:-------------:|:------------------------:|:---------------:|:----------:|
62
+ | 0.2546 | 1.0 | 343 | 0.3220 | 0.5965 | 0.6868 | 0.8757 | 0.8517 | 0.9218 | 0.7201 | 0.0 | 0.9404 | 0.7384 | 0.8585 | 0.5502 | 0.0 | 0.8353 |
63
+ | 0.336 | 2.0 | 686 | 0.3159 | 0.5992 | 0.6766 | 0.8807 | 0.8816 | 0.9295 | 0.6220 | 0.0 | 0.9497 | 0.7474 | 0.8627 | 0.5429 | 0.0 | 0.8428 |
64
+ | 0.2976 | 3.0 | 1029 | 0.3087 | 0.6057 | 0.6971 | 0.8807 | 0.8383 | 0.9325 | 0.7561 | 0.0000 | 0.9583 | 0.7412 | 0.8629 | 0.5833 | 0.0000 | 0.8411 |
65
+ | 0.2791 | 4.0 | 1372 | 0.2907 | 0.6175 | 0.6995 | 0.8886 | 0.8717 | 0.9290 | 0.7401 | 0.0016 | 0.9548 | 0.7608 | 0.8674 | 0.6070 | 0.0016 | 0.8507 |
66
+ | 0.2795 | 5.0 | 1715 | 0.2883 | 0.6264 | 0.7025 | 0.8906 | 0.8675 | 0.9369 | 0.7303 | 0.0291 | 0.9489 | 0.7630 | 0.8689 | 0.6135 | 0.0283 | 0.8584 |
67
+ | 0.2215 | 6.0 | 2058 | 0.2845 | 0.6316 | 0.7081 | 0.8924 | 0.8873 | 0.9252 | 0.7457 | 0.0452 | 0.9373 | 0.7700 | 0.8700 | 0.6212 | 0.0431 | 0.8536 |
68
+ | 0.2372 | 7.0 | 2401 | 0.2770 | 0.6343 | 0.7197 | 0.8931 | 0.8565 | 0.9373 | 0.7906 | 0.0492 | 0.9651 | 0.7657 | 0.8715 | 0.6365 | 0.0472 | 0.8504 |
69
+ | 0.3055 | 8.0 | 2744 | 0.2742 | 0.6337 | 0.7201 | 0.8950 | 0.8835 | 0.9220 | 0.8026 | 0.0324 | 0.9603 | 0.7742 | 0.8728 | 0.6413 | 0.0317 | 0.8482 |
70
+ | 0.2047 | 9.0 | 3087 | 0.2680 | 0.6497 | 0.7251 | 0.8982 | 0.8733 | 0.9384 | 0.7786 | 0.0884 | 0.9468 | 0.7765 | 0.8766 | 0.6500 | 0.0819 | 0.8634 |
71
+ | 0.1705 | 10.0 | 3430 | 0.2675 | 0.6489 | 0.7328 | 0.8987 | 0.8744 | 0.9336 | 0.8043 | 0.0862 | 0.9654 | 0.7793 | 0.8767 | 0.6531 | 0.0802 | 0.8550 |
72
+ | 0.2029 | 11.0 | 3773 | 0.2685 | 0.6523 | 0.7267 | 0.9003 | 0.8751 | 0.9443 | 0.7596 | 0.0958 | 0.9589 | 0.7812 | 0.8779 | 0.6536 | 0.0890 | 0.8600 |
73
+ | 0.1707 | 12.0 | 4116 | 0.2612 | 0.6591 | 0.7360 | 0.9015 | 0.8866 | 0.9324 | 0.7982 | 0.1097 | 0.9532 | 0.7853 | 0.8788 | 0.6639 | 0.0995 | 0.8679 |
74
+ | 0.2742 | 13.0 | 4459 | 0.2628 | 0.6512 | 0.7247 | 0.9022 | 0.8756 | 0.9442 | 0.7781 | 0.0666 | 0.9593 | 0.7847 | 0.8797 | 0.6635 | 0.0633 | 0.8651 |
75
+ | 0.2991 | 14.0 | 4802 | 0.2702 | 0.6653 | 0.7404 | 0.9025 | 0.8909 | 0.9368 | 0.7673 | 0.1492 | 0.9578 | 0.7870 | 0.8799 | 0.6627 | 0.1247 | 0.8722 |
76
+ | 0.229 | 15.0 | 5145 | 0.2599 | 0.6615 | 0.7395 | 0.9026 | 0.8723 | 0.9463 | 0.7800 | 0.1303 | 0.9687 | 0.7850 | 0.8798 | 0.6682 | 0.1143 | 0.8604 |
77
+ | 0.2004 | 16.0 | 5488 | 0.2595 | 0.6719 | 0.7473 | 0.9042 | 0.8854 | 0.9398 | 0.7863 | 0.1735 | 0.9513 | 0.7898 | 0.8814 | 0.6719 | 0.1442 | 0.8721 |
78
+ | 0.1944 | 17.0 | 5831 | 0.2564 | 0.6729 | 0.7486 | 0.9058 | 0.8940 | 0.9368 | 0.7895 | 0.1693 | 0.9536 | 0.7936 | 0.8830 | 0.6778 | 0.1418 | 0.8685 |
79
+ | 0.2068 | 18.0 | 6174 | 0.2539 | 0.6664 | 0.7450 | 0.9061 | 0.8915 | 0.9362 | 0.8051 | 0.1245 | 0.9677 | 0.7940 | 0.8839 | 0.6801 | 0.1102 | 0.8641 |
80
+ | 0.2461 | 19.0 | 6517 | 0.2494 | 0.6776 | 0.7603 | 0.9063 | 0.8756 | 0.9427 | 0.8251 | 0.1941 | 0.9642 | 0.7927 | 0.8854 | 0.6800 | 0.1585 | 0.8712 |
81
+ | 0.2252 | 20.0 | 6860 | 0.2498 | 0.6733 | 0.7461 | 0.9074 | 0.8813 | 0.9452 | 0.8043 | 0.1456 | 0.9542 | 0.7947 | 0.8856 | 0.6843 | 0.1284 | 0.8736 |
82
+ | 0.1975 | 21.0 | 7203 | 0.2519 | 0.6761 | 0.7516 | 0.9084 | 0.8960 | 0.9386 | 0.7992 | 0.1656 | 0.9585 | 0.7989 | 0.8861 | 0.6862 | 0.1412 | 0.8679 |
83
+ | 0.2356 | 22.0 | 7546 | 0.2506 | 0.6801 | 0.7526 | 0.9087 | 0.8956 | 0.9396 | 0.7972 | 0.1764 | 0.9542 | 0.7991 | 0.8858 | 0.6890 | 0.1486 | 0.8779 |
84
+ | 0.1838 | 23.0 | 7889 | 0.2510 | 0.6805 | 0.7554 | 0.9088 | 0.8835 | 0.9455 | 0.8068 | 0.1824 | 0.9589 | 0.7978 | 0.8867 | 0.6892 | 0.1516 | 0.8773 |
85
+ | 0.1576 | 24.0 | 8232 | 0.2511 | 0.6850 | 0.7658 | 0.9091 | 0.8913 | 0.9418 | 0.8021 | 0.2291 | 0.9650 | 0.7996 | 0.8868 | 0.6891 | 0.1765 | 0.8731 |
86
+ | 0.1504 | 25.0 | 8575 | 0.2505 | 0.6819 | 0.7590 | 0.9092 | 0.8869 | 0.9439 | 0.8077 | 0.1916 | 0.9650 | 0.7992 | 0.8873 | 0.6890 | 0.1587 | 0.8751 |
87
+ | 0.2196 | 26.0 | 8918 | 0.2530 | 0.6830 | 0.7597 | 0.9095 | 0.8946 | 0.9405 | 0.8035 | 0.1985 | 0.9612 | 0.8010 | 0.8872 | 0.6900 | 0.1610 | 0.8756 |
88
+ | 0.1781 | 27.0 | 9261 | 0.2509 | 0.6841 | 0.7596 | 0.9101 | 0.8901 | 0.9451 | 0.7993 | 0.2000 | 0.9635 | 0.8010 | 0.8880 | 0.6930 | 0.1635 | 0.8749 |
89
+ | 0.1578 | 28.0 | 9604 | 0.2485 | 0.6831 | 0.7591 | 0.9102 | 0.8874 | 0.9457 | 0.8064 | 0.1912 | 0.9651 | 0.8008 | 0.8882 | 0.6942 | 0.1585 | 0.8740 |
90
+ | 0.1931 | 29.0 | 9947 | 0.2495 | 0.6840 | 0.7579 | 0.9105 | 0.8893 | 0.9454 | 0.8042 | 0.1899 | 0.9604 | 0.8016 | 0.8884 | 0.6940 | 0.1580 | 0.8779 |
91
+ | 0.1582 | 30.0 | 10290 | 0.2486 | 0.6867 | 0.7623 | 0.9106 | 0.8910 | 0.9442 | 0.8061 | 0.2109 | 0.9591 | 0.8022 | 0.8886 | 0.6946 | 0.1697 | 0.8786 |
92
+
93
+
94
+ ### Framework versions
95
+
96
+ - Transformers 4.33.2
97
+ - Pytorch 2.0.1+cu118
98
+ - Datasets 2.14.5
99
+ - Tokenizers 0.13.3
config.json ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "nvidia/mit-b0",
3
+ "architectures": [
4
+ "SegformerForSemanticSegmentation"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "classifier_dropout_prob": 0.1,
8
+ "decoder_hidden_size": 256,
9
+ "depths": [
10
+ 2,
11
+ 2,
12
+ 2,
13
+ 2
14
+ ],
15
+ "downsampling_rates": [
16
+ 1,
17
+ 4,
18
+ 8,
19
+ 16
20
+ ],
21
+ "drop_path_rate": 0.1,
22
+ "hidden_act": "gelu",
23
+ "hidden_dropout_prob": 0.0,
24
+ "hidden_sizes": [
25
+ 32,
26
+ 64,
27
+ 160,
28
+ 256
29
+ ],
30
+ "id2label": {
31
+ "0": "Matrix",
32
+ "1": "Austenite",
33
+ "2": "Martensite/Austenite",
34
+ "3": "Precipitate",
35
+ "4": "Defect"
36
+ },
37
+ "image_size": 224,
38
+ "initializer_range": 0.02,
39
+ "label2id": {
40
+ "Austenite": 1,
41
+ "Defect": 4,
42
+ "Martensite/Austenite": 2,
43
+ "Matrix": 0,
44
+ "Precipitate": 3
45
+ },
46
+ "layer_norm_eps": 1e-06,
47
+ "mlp_ratios": [
48
+ 4,
49
+ 4,
50
+ 4,
51
+ 4
52
+ ],
53
+ "model_type": "segformer",
54
+ "num_attention_heads": [
55
+ 1,
56
+ 2,
57
+ 5,
58
+ 8
59
+ ],
60
+ "num_channels": 3,
61
+ "num_encoder_blocks": 4,
62
+ "patch_sizes": [
63
+ 7,
64
+ 3,
65
+ 3,
66
+ 3
67
+ ],
68
+ "reshape_last_stage": true,
69
+ "semantic_loss_ignore_index": 255,
70
+ "sr_ratios": [
71
+ 8,
72
+ 4,
73
+ 2,
74
+ 1
75
+ ],
76
+ "strides": [
77
+ 4,
78
+ 2,
79
+ 2,
80
+ 2
81
+ ],
82
+ "torch_dtype": "float32",
83
+ "transformers_version": "4.33.2"
84
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ba026266c0d769d8ad8c034c789d4c71e3c77a0c8b8126bf5c3602e0d86e232
3
+ size 14934861
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1475f80ec8fb284dd130d1a3f26f3828935f068b3dd01e48a60687c223319bba
3
+ size 4091