lwe0105 commited on
Commit
958b9f8
·
verified ·
1 Parent(s): 43dacee

Model save

Browse files
README.md ADDED
@@ -0,0 +1,112 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: other
4
+ base_model: nvidia/mit-b1
5
+ tags:
6
+ - generated_from_trainer
7
+ model-index:
8
+ - name: segformer-b1-GFB
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # segformer-b1-GFB
16
+
17
+ This model is a fine-tuned version of [nvidia/mit-b1](https://huggingface.co/nvidia/mit-b1) on the None dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.4850
20
+ - Mean Iou: 0.6288
21
+ - Mean Accuracy: 0.7339
22
+ - Overall Accuracy: 0.8955
23
+ - Accuracy Unlabeled: 0.9551
24
+ - Accuracy Gbm: 0.7834
25
+ - Accuracy Podo: 0.6686
26
+ - Accuracy Endo: 0.5283
27
+ - Iou Unlabeled: 0.8942
28
+ - Iou Gbm: 0.6485
29
+ - Iou Podo: 0.5402
30
+ - Iou Endo: 0.4322
31
+
32
+ ## Model description
33
+
34
+ More information needed
35
+
36
+ ## Intended uses & limitations
37
+
38
+ More information needed
39
+
40
+ ## Training and evaluation data
41
+
42
+ More information needed
43
+
44
+ ## Training procedure
45
+
46
+ ### Training hyperparameters
47
+
48
+ The following hyperparameters were used during training:
49
+ - learning_rate: 0.0002
50
+ - train_batch_size: 32
51
+ - eval_batch_size: 32
52
+ - seed: 42
53
+ - optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
54
+ - lr_scheduler_type: cosine
55
+ - lr_scheduler_warmup_steps: 400
56
+ - num_epochs: 100
57
+
58
+ ### Training results
59
+
60
+ | Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy Unlabeled | Accuracy Gbm | Accuracy Podo | Accuracy Endo | Iou Unlabeled | Iou Gbm | Iou Podo | Iou Endo |
61
+ |:-------------:|:-------:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:------------------:|:------------:|:-------------:|:-------------:|:-------------:|:-------:|:--------:|:--------:|
62
+ | 0.8169 | 2.3256 | 100 | 0.7654 | 0.3767 | 0.4605 | 0.8186 | 0.9525 | 0.6268 | 0.1952 | 0.0676 | 0.8295 | 0.4619 | 0.1615 | 0.0542 |
63
+ | 0.4671 | 4.6512 | 200 | 0.4801 | 0.4802 | 0.5915 | 0.8530 | 0.9370 | 0.7438 | 0.5501 | 0.1352 | 0.8635 | 0.5637 | 0.3742 | 0.1194 |
64
+ | 0.3704 | 6.9767 | 300 | 0.3731 | 0.5255 | 0.6070 | 0.8701 | 0.9686 | 0.6797 | 0.4928 | 0.2869 | 0.8707 | 0.5734 | 0.4128 | 0.2451 |
65
+ | 0.323 | 9.3023 | 400 | 0.3512 | 0.5717 | 0.7089 | 0.8704 | 0.9219 | 0.8538 | 0.6057 | 0.4541 | 0.8732 | 0.5957 | 0.4614 | 0.3564 |
66
+ | 0.2314 | 11.6279 | 500 | 0.3261 | 0.5873 | 0.6924 | 0.8830 | 0.9524 | 0.7566 | 0.6179 | 0.4425 | 0.8837 | 0.6130 | 0.4872 | 0.3654 |
67
+ | 0.1931 | 13.9535 | 600 | 0.3339 | 0.5970 | 0.6926 | 0.8879 | 0.9607 | 0.7528 | 0.6018 | 0.4549 | 0.8881 | 0.6189 | 0.4972 | 0.3839 |
68
+ | 0.1867 | 16.2791 | 700 | 0.3734 | 0.5898 | 0.6759 | 0.8861 | 0.9677 | 0.7028 | 0.5926 | 0.4406 | 0.8845 | 0.6059 | 0.4917 | 0.3770 |
69
+ | 0.1687 | 18.6047 | 800 | 0.3497 | 0.6126 | 0.7306 | 0.8879 | 0.9449 | 0.7979 | 0.6507 | 0.5290 | 0.8874 | 0.6321 | 0.5155 | 0.4155 |
70
+ | 0.1533 | 20.9302 | 900 | 0.3721 | 0.6120 | 0.7204 | 0.8889 | 0.9534 | 0.7442 | 0.6642 | 0.5200 | 0.8875 | 0.6240 | 0.5243 | 0.4121 |
71
+ | 0.151 | 23.2558 | 1000 | 0.4038 | 0.6056 | 0.7027 | 0.8890 | 0.9617 | 0.7033 | 0.6609 | 0.4851 | 0.8869 | 0.6146 | 0.5264 | 0.3945 |
72
+ | 0.1331 | 25.5814 | 1100 | 0.3953 | 0.6115 | 0.7087 | 0.8914 | 0.9625 | 0.7503 | 0.6029 | 0.5192 | 0.8905 | 0.6277 | 0.5101 | 0.4175 |
73
+ | 0.1111 | 27.9070 | 1200 | 0.3559 | 0.6246 | 0.7452 | 0.8911 | 0.9416 | 0.8125 | 0.6980 | 0.5286 | 0.8904 | 0.6356 | 0.5435 | 0.4287 |
74
+ | 0.1032 | 30.2326 | 1300 | 0.3800 | 0.6210 | 0.7438 | 0.8895 | 0.9385 | 0.8257 | 0.6939 | 0.5171 | 0.8884 | 0.6362 | 0.5388 | 0.4205 |
75
+ | 0.103 | 32.5581 | 1400 | 0.3927 | 0.6210 | 0.7287 | 0.8928 | 0.9533 | 0.7868 | 0.6498 | 0.5251 | 0.8917 | 0.6414 | 0.5294 | 0.4215 |
76
+ | 0.0954 | 34.8837 | 1500 | 0.3889 | 0.6230 | 0.7373 | 0.8922 | 0.9463 | 0.8084 | 0.6814 | 0.5132 | 0.8910 | 0.6431 | 0.5393 | 0.4188 |
77
+ | 0.1129 | 37.2093 | 1600 | 0.4255 | 0.6187 | 0.7250 | 0.8916 | 0.9559 | 0.7465 | 0.6649 | 0.5328 | 0.8904 | 0.6308 | 0.5292 | 0.4244 |
78
+ | 0.1053 | 39.5349 | 1700 | 0.4003 | 0.6238 | 0.7298 | 0.8943 | 0.9540 | 0.7867 | 0.6673 | 0.5113 | 0.8937 | 0.6448 | 0.5362 | 0.4204 |
79
+ | 0.0748 | 41.8605 | 1800 | 0.4021 | 0.6290 | 0.7416 | 0.8945 | 0.9506 | 0.7862 | 0.6877 | 0.5419 | 0.8936 | 0.6479 | 0.5412 | 0.4333 |
80
+ | 0.1041 | 44.1860 | 1900 | 0.4126 | 0.6239 | 0.7270 | 0.8949 | 0.9557 | 0.7902 | 0.6569 | 0.5051 | 0.8939 | 0.6477 | 0.5355 | 0.4186 |
81
+ | 0.0876 | 46.5116 | 2000 | 0.4366 | 0.6226 | 0.7210 | 0.8948 | 0.9600 | 0.7599 | 0.6571 | 0.5071 | 0.8929 | 0.6425 | 0.5374 | 0.4173 |
82
+ | 0.0948 | 48.8372 | 2100 | 0.4274 | 0.6196 | 0.7162 | 0.8944 | 0.9615 | 0.7581 | 0.6458 | 0.4993 | 0.8931 | 0.6407 | 0.5313 | 0.4133 |
83
+ | 0.0755 | 51.1628 | 2200 | 0.4197 | 0.6287 | 0.7427 | 0.8939 | 0.9487 | 0.8031 | 0.6725 | 0.5465 | 0.8925 | 0.6475 | 0.5395 | 0.4355 |
84
+ | 0.0825 | 53.4884 | 2300 | 0.4488 | 0.6248 | 0.7289 | 0.8945 | 0.9567 | 0.7708 | 0.6592 | 0.5289 | 0.8935 | 0.6413 | 0.5344 | 0.4301 |
85
+ | 0.0871 | 55.8140 | 2400 | 0.4385 | 0.6295 | 0.7401 | 0.8948 | 0.9506 | 0.7948 | 0.6837 | 0.5313 | 0.8937 | 0.6482 | 0.5430 | 0.4329 |
86
+ | 0.0808 | 58.1395 | 2500 | 0.4318 | 0.6308 | 0.7448 | 0.8944 | 0.9486 | 0.7980 | 0.6871 | 0.5454 | 0.8930 | 0.6462 | 0.5461 | 0.4380 |
87
+ | 0.0764 | 60.4651 | 2600 | 0.4608 | 0.6273 | 0.7347 | 0.8945 | 0.9537 | 0.7836 | 0.6658 | 0.5356 | 0.8930 | 0.6464 | 0.5363 | 0.4334 |
88
+ | 0.0806 | 62.7907 | 2700 | 0.4616 | 0.6272 | 0.7380 | 0.8937 | 0.9508 | 0.7877 | 0.6774 | 0.5362 | 0.8923 | 0.6449 | 0.5384 | 0.4335 |
89
+ | 0.0785 | 65.1163 | 2800 | 0.4292 | 0.6292 | 0.7538 | 0.8918 | 0.9400 | 0.8149 | 0.7057 | 0.5547 | 0.8905 | 0.6471 | 0.5393 | 0.4400 |
90
+ | 0.0591 | 67.4419 | 2900 | 0.4719 | 0.6266 | 0.7312 | 0.8948 | 0.9550 | 0.7761 | 0.6757 | 0.5180 | 0.8933 | 0.6459 | 0.5401 | 0.4270 |
91
+ | 0.0723 | 69.7674 | 3000 | 0.4687 | 0.6289 | 0.7363 | 0.8951 | 0.9537 | 0.7767 | 0.6850 | 0.5299 | 0.8939 | 0.6452 | 0.5445 | 0.4322 |
92
+ | 0.0774 | 72.0930 | 3100 | 0.4575 | 0.6293 | 0.7367 | 0.8954 | 0.9532 | 0.7885 | 0.6784 | 0.5267 | 0.8942 | 0.6487 | 0.5432 | 0.4309 |
93
+ | 0.0785 | 74.4186 | 3200 | 0.4689 | 0.6277 | 0.7364 | 0.8945 | 0.9521 | 0.7891 | 0.6767 | 0.5280 | 0.8932 | 0.6478 | 0.5399 | 0.4299 |
94
+ | 0.0572 | 76.7442 | 3300 | 0.4598 | 0.6323 | 0.7465 | 0.8950 | 0.9496 | 0.7999 | 0.6766 | 0.5600 | 0.8939 | 0.6498 | 0.5423 | 0.4431 |
95
+ | 0.0852 | 79.0698 | 3400 | 0.4713 | 0.6305 | 0.7393 | 0.8954 | 0.9530 | 0.7836 | 0.6815 | 0.5392 | 0.8941 | 0.6479 | 0.5441 | 0.4360 |
96
+ | 0.0871 | 81.3953 | 3500 | 0.4742 | 0.6291 | 0.7366 | 0.8951 | 0.9539 | 0.7858 | 0.6678 | 0.5390 | 0.8939 | 0.6486 | 0.5379 | 0.4361 |
97
+ | 0.0589 | 83.7209 | 3600 | 0.4749 | 0.6292 | 0.7371 | 0.8952 | 0.9526 | 0.7939 | 0.6720 | 0.5300 | 0.8938 | 0.6502 | 0.5397 | 0.4330 |
98
+ | 0.0621 | 86.0465 | 3700 | 0.4755 | 0.6301 | 0.7381 | 0.8955 | 0.9530 | 0.7913 | 0.6744 | 0.5335 | 0.8943 | 0.6497 | 0.5419 | 0.4345 |
99
+ | 0.0649 | 88.3721 | 3800 | 0.4855 | 0.6289 | 0.7361 | 0.8951 | 0.9538 | 0.7855 | 0.6707 | 0.5343 | 0.8936 | 0.6478 | 0.5403 | 0.4339 |
100
+ | 0.0591 | 90.6977 | 3900 | 0.4811 | 0.6290 | 0.7356 | 0.8953 | 0.9537 | 0.7867 | 0.6746 | 0.5273 | 0.8939 | 0.6488 | 0.5414 | 0.4319 |
101
+ | 0.0584 | 93.0233 | 4000 | 0.4799 | 0.6303 | 0.7382 | 0.8955 | 0.9532 | 0.7865 | 0.6786 | 0.5347 | 0.8941 | 0.6487 | 0.5432 | 0.4351 |
102
+ | 0.0615 | 95.3488 | 4100 | 0.4847 | 0.6295 | 0.7364 | 0.8955 | 0.9541 | 0.7858 | 0.6713 | 0.5342 | 0.8941 | 0.6487 | 0.5410 | 0.4343 |
103
+ | 0.0596 | 97.6744 | 4200 | 0.4862 | 0.6288 | 0.7339 | 0.8956 | 0.9552 | 0.7837 | 0.6675 | 0.5291 | 0.8942 | 0.6485 | 0.5400 | 0.4325 |
104
+ | 0.0603 | 100.0 | 4300 | 0.4850 | 0.6288 | 0.7339 | 0.8955 | 0.9551 | 0.7834 | 0.6686 | 0.5283 | 0.8942 | 0.6485 | 0.5402 | 0.4322 |
105
+
106
+
107
+ ### Framework versions
108
+
109
+ - Transformers 4.57.2
110
+ - Pytorch 2.8.0+cu126
111
+ - Datasets 4.4.1
112
+ - Tokenizers 0.22.1
config.json ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SegformerForSemanticSegmentation"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.0,
6
+ "classifier_dropout_prob": 0.1,
7
+ "decoder_hidden_size": 256,
8
+ "depths": [
9
+ 2,
10
+ 2,
11
+ 2,
12
+ 2
13
+ ],
14
+ "downsampling_rates": [
15
+ 1,
16
+ 4,
17
+ 8,
18
+ 16
19
+ ],
20
+ "drop_path_rate": 0.1,
21
+ "dtype": "float32",
22
+ "hidden_act": "gelu",
23
+ "hidden_dropout_prob": 0.0,
24
+ "hidden_sizes": [
25
+ 64,
26
+ 128,
27
+ 320,
28
+ 512
29
+ ],
30
+ "id2label": {
31
+ "0": "unlabeled",
32
+ "1": "GBM",
33
+ "2": "Podo",
34
+ "3": "Endo"
35
+ },
36
+ "image_size": 224,
37
+ "initializer_range": 0.02,
38
+ "label2id": {
39
+ "Endo": 3,
40
+ "GBM": 1,
41
+ "Podo": 2,
42
+ "unlabeled": 0
43
+ },
44
+ "layer_norm_eps": 1e-06,
45
+ "mlp_ratios": [
46
+ 4,
47
+ 4,
48
+ 4,
49
+ 4
50
+ ],
51
+ "model_type": "segformer",
52
+ "num_attention_heads": [
53
+ 1,
54
+ 2,
55
+ 5,
56
+ 8
57
+ ],
58
+ "num_channels": 3,
59
+ "num_encoder_blocks": 4,
60
+ "patch_sizes": [
61
+ 7,
62
+ 3,
63
+ 3,
64
+ 3
65
+ ],
66
+ "reshape_last_stage": true,
67
+ "semantic_loss_ignore_index": 255,
68
+ "sr_ratios": [
69
+ 8,
70
+ 4,
71
+ 2,
72
+ 1
73
+ ],
74
+ "strides": [
75
+ 4,
76
+ 2,
77
+ 2,
78
+ 2
79
+ ],
80
+ "transformers_version": "4.57.2"
81
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b801023f2592dbcf86120d9b5e3155fb4e4eeab68b3472b719c866a10b9fa3a
3
+ size 54739432
runs/Nov26_11-20-53_bio-Super-Server/events.out.tfevents.1764127258.bio-Super-Server.2341633.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57b962e08783cf682096cc50efa8e6e819e080860df7dee38e176dc4aa166d12
3
+ size 92968
runs/Nov26_11-48-01_bio-Super-Server/events.out.tfevents.1764128911.bio-Super-Server.2357102.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c27efb3463096d534499e94997fe6daf02e9a36a9348d67d59b1cf9a078da9ec
3
+ size 5556
runs/Nov26_11-51-55_bio-Super-Server/events.out.tfevents.1764129119.bio-Super-Server.2357102.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:081a30570ea626eae3679860b5c4067cea70c99c3240425e4d1b8bee4d9a141f
3
+ size 5556
runs/Nov26_11-54-59_bio-Super-Server/events.out.tfevents.1764129303.bio-Super-Server.2357102.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:619e59fbc26f94ac50c96ebec0b6ac6eb9e3c672ae9df799f2aea5141752d460
3
+ size 5556
runs/Nov26_11-56-30_bio-Super-Server/events.out.tfevents.1764129394.bio-Super-Server.2357102.3 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bca985d7fbf603193f0ed557d790e892301146dfca208b2760df5fcbb659196b
3
+ size 5556
runs/Nov26_13-12-18_bio-Super-Server/events.out.tfevents.1764133944.bio-Super-Server.2430896.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2f9c358acf94197616944df0d4196cecc2eb9a12d5aeb0243d2b33c22c3eeb7
3
+ size 5556
runs/Nov26_13-16-19_bio-Super-Server/events.out.tfevents.1764134183.bio-Super-Server.2433711.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44cdb1cbd10f2539a2f6e0caeb56c74541a3e1bd8185846bba0c7d1c7b0b8e1c
3
+ size 125502
runs/Nov26_14-06-42_bio-Super-Server/events.out.tfevents.1764137208.bio-Super-Server.2451216.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66e46f1d616b91899f36d8c7a58d8759ab59f799ed454c60a80d7b1bdb924596
3
+ size 18183
runs/Nov26_14-20-02_bio-Super-Server/events.out.tfevents.1764138013.bio-Super-Server.2451216.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d5a554974fe43ad72e07107490c358a169e701f2a80f5755ca0cd2bdd4d5c4d
3
+ size 65035
runs/Nov26_15-51-20_bio-Super-Server/events.out.tfevents.1764143593.bio-Super-Server.2543904.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a031af48cf88c8f9692c73912d86a78362f34f114935b33cf25c1b48ebdc25a2
3
+ size 231835
runs/Nov26_16-50-25_bio-Super-Server/events.out.tfevents.1764147145.bio-Super-Server.2634446.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a22847b9c284a2f9f304f94364c7624e3e741baeb0106cfb9e29df114b6c6083
3
+ size 48116
runs/Nov26_17-00-19_bio-Super-Server/events.out.tfevents.1764147626.bio-Super-Server.2648179.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f3e7af65f0f9f0d82edcea86f9d7236399c92f22b2503f0d3cf0dbb48b7469d
3
+ size 950701
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a7aa365c211e086bc5d84540ded83219444106450e6f5b4231714d57d305972
3
+ size 5905