Hugging Face's logo

matthieulel
/
dinov2-small-imagenet1k-1-layer-finetuned-galaxy10-decals

Image Classification
Transformers
TensorBoard
Safetensors
dinov2
vision
Generated from Trainer
Model card Files Metrics Community
matthieulel commited on
Commit
066b8f3
·
verified ·
1 Parent(s): 4ffdd47

Model save

Browse files
Files changed (2) hide show
  1. README.md +48 -34
  2. model.safetensors +1 -1
README.md CHANGED
@@ -2,11 +2,12 @@
2
  license: apache-2.0
3
  base_model: facebook/dinov2-small-imagenet1k-1-layer
4
  tags:
5
- - image-classification
6
- - vision
7
  - generated_from_trainer
8
  metrics:
9
  - accuracy
 
 
 
10
  model-index:
11
  - name: dinov2-small-imagenet1k-1-layer-finetuned-galaxy10-decals
12
  results: []
@@ -17,10 +18,13 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  # dinov2-small-imagenet1k-1-layer-finetuned-galaxy10-decals
19
 
20
- This model is a fine-tuned version of [facebook/dinov2-small-imagenet1k-1-layer](https://huggingface.co/facebook/dinov2-small-imagenet1k-1-layer) on the matthieulel/galaxy10_decals dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.4242
23
- - Accuracy: 0.8715
 
 
 
24
 
25
  ## Model description
26
 
@@ -40,45 +44,55 @@ More information needed
40
 
41
  The following hyperparameters were used during training:
42
  - learning_rate: 5e-05
43
- - train_batch_size: 32
44
- - eval_batch_size: 32
45
  - seed: 42
46
  - gradient_accumulation_steps: 4
47
- - total_train_batch_size: 128
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
  - lr_scheduler_warmup_ratio: 0.1
51
- - num_epochs: 20
52
 
53
  ### Training results
54
 
55
- | Training Loss | Epoch | Step | Validation Loss | Accuracy |
56
- |:-------------:|:-------:|:----:|:---------------:|:--------:|
57
- | 0.924 | 0.9940 | 124 | 0.8075 | 0.7294 |
58
- | 0.8226 | 1.9960 | 249 | 0.8195 | 0.7193 |
59
- | 0.805 | 2.9980 | 374 | 0.6233 | 0.7903 |
60
- | 0.752 | 4.0 | 499 | 0.6687 | 0.7796 |
61
- | 0.6193 | 4.9940 | 623 | 0.5708 | 0.8050 |
62
- | 0.5822 | 5.9960 | 748 | 0.5624 | 0.8061 |
63
- | 0.6227 | 6.9980 | 873 | 0.5536 | 0.8134 |
64
- | 0.6067 | 8.0 | 998 | 0.5687 | 0.8185 |
65
- | 0.533 | 8.9940 | 1122 | 0.5919 | 0.8089 |
66
- | 0.5423 | 9.9960 | 1247 | 0.5022 | 0.8371 |
67
- | 0.4747 | 10.9980 | 1372 | 0.4419 | 0.8579 |
68
- | 0.4367 | 12.0 | 1497 | 0.5084 | 0.8207 |
69
- | 0.4907 | 12.9940 | 1621 | 0.4774 | 0.8365 |
70
- | 0.4269 | 13.9960 | 1746 | 0.4945 | 0.8444 |
71
- | 0.3787 | 14.9980 | 1871 | 0.4779 | 0.8478 |
72
- | 0.3724 | 16.0 | 1996 | 0.4242 | 0.8715 |
73
- | 0.354 | 16.9940 | 2120 | 0.4595 | 0.8529 |
74
- | 0.3304 | 17.9960 | 2245 | 0.4543 | 0.8563 |
75
- | 0.2849 | 18.9980 | 2370 | 0.4699 | 0.8608 |
76
- | 0.2456 | 19.8798 | 2480 | 0.4505 | 0.8664 |
 
 
 
 
 
 
 
 
 
 
77
 
78
 
79
  ### Framework versions
80
 
81
- - Transformers 4.40.1
82
- - Pytorch 2.2.1+cu121
83
  - Datasets 2.19.1
84
- - Tokenizers 0.19.1
 
2
  license: apache-2.0
3
  base_model: facebook/dinov2-small-imagenet1k-1-layer
4
  tags:
 
 
5
  - generated_from_trainer
6
  metrics:
7
  - accuracy
8
+ - precision
9
+ - recall
10
+ - f1
11
  model-index:
12
  - name: dinov2-small-imagenet1k-1-layer-finetuned-galaxy10-decals
13
  results: []
 
18
 
19
  # dinov2-small-imagenet1k-1-layer-finetuned-galaxy10-decals
20
 
21
+ This model is a fine-tuned version of [facebook/dinov2-small-imagenet1k-1-layer](https://huggingface.co/facebook/dinov2-small-imagenet1k-1-layer) on an unknown dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.5429
24
+ - Accuracy: 0.8512
25
+ - Precision: 0.8489
26
+ - Recall: 0.8512
27
+ - F1: 0.8495
28
 
29
  ## Model description
30
 
 
44
 
45
  The following hyperparameters were used during training:
46
  - learning_rate: 5e-05
47
+ - train_batch_size: 64
48
+ - eval_batch_size: 64
49
  - seed: 42
50
  - gradient_accumulation_steps: 4
51
+ - total_train_batch_size: 256
52
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
  - lr_scheduler_type: linear
54
  - lr_scheduler_warmup_ratio: 0.1
55
+ - num_epochs: 30
56
 
57
  ### Training results
58
 
59
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
60
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
61
+ | 1.0027 | 0.99 | 62 | 0.8262 | 0.7080 | 0.7231 | 0.7080 | 0.7017 |
62
+ | 0.8374 | 2.0 | 125 | 0.6129 | 0.7948 | 0.7960 | 0.7948 | 0.7899 |
63
+ | 0.7753 | 2.99 | 187 | 0.6555 | 0.7892 | 0.7921 | 0.7892 | 0.7787 |
64
+ | 0.7165 | 4.0 | 250 | 0.5862 | 0.8005 | 0.8053 | 0.8005 | 0.7970 |
65
+ | 0.6477 | 4.99 | 312 | 0.6183 | 0.7965 | 0.8119 | 0.7965 | 0.7985 |
66
+ | 0.6892 | 6.0 | 375 | 0.5310 | 0.8247 | 0.8275 | 0.8247 | 0.8195 |
67
+ | 0.6171 | 6.99 | 437 | 0.5678 | 0.8083 | 0.8157 | 0.8083 | 0.8022 |
68
+ | 0.55 | 8.0 | 500 | 0.4961 | 0.8326 | 0.8353 | 0.8326 | 0.8316 |
69
+ | 0.5615 | 8.99 | 562 | 0.5033 | 0.8309 | 0.8312 | 0.8309 | 0.8274 |
70
+ | 0.5107 | 10.0 | 625 | 0.5162 | 0.8191 | 0.8164 | 0.8191 | 0.8152 |
71
+ | 0.5237 | 10.99 | 687 | 0.4790 | 0.8422 | 0.8452 | 0.8422 | 0.8381 |
72
+ | 0.4954 | 12.0 | 750 | 0.4782 | 0.8422 | 0.8430 | 0.8422 | 0.8373 |
73
+ | 0.4887 | 12.99 | 812 | 0.4689 | 0.8371 | 0.8395 | 0.8371 | 0.8358 |
74
+ | 0.4629 | 14.0 | 875 | 0.4541 | 0.8523 | 0.8500 | 0.8523 | 0.8502 |
75
+ | 0.4486 | 14.99 | 937 | 0.4755 | 0.8405 | 0.8400 | 0.8405 | 0.8394 |
76
+ | 0.4361 | 16.0 | 1000 | 0.4763 | 0.8371 | 0.8392 | 0.8371 | 0.8370 |
77
+ | 0.3833 | 16.99 | 1062 | 0.4982 | 0.8416 | 0.8429 | 0.8416 | 0.8396 |
78
+ | 0.3788 | 18.0 | 1125 | 0.5632 | 0.8292 | 0.8365 | 0.8292 | 0.8267 |
79
+ | 0.3722 | 18.99 | 1187 | 0.5162 | 0.8388 | 0.8364 | 0.8388 | 0.8357 |
80
+ | 0.3467 | 20.0 | 1250 | 0.5125 | 0.8399 | 0.8357 | 0.8399 | 0.8342 |
81
+ | 0.3518 | 20.99 | 1312 | 0.5569 | 0.8309 | 0.8327 | 0.8309 | 0.8276 |
82
+ | 0.3432 | 22.0 | 1375 | 0.5032 | 0.8484 | 0.8451 | 0.8484 | 0.8454 |
83
+ | 0.3067 | 22.99 | 1437 | 0.5246 | 0.8433 | 0.8462 | 0.8433 | 0.8433 |
84
+ | 0.2923 | 24.0 | 1500 | 0.5363 | 0.8467 | 0.8482 | 0.8467 | 0.8464 |
85
+ | 0.303 | 24.99 | 1562 | 0.5435 | 0.8484 | 0.8453 | 0.8484 | 0.8457 |
86
+ | 0.2523 | 26.0 | 1625 | 0.5500 | 0.8444 | 0.8422 | 0.8444 | 0.8419 |
87
+ | 0.2523 | 26.99 | 1687 | 0.5369 | 0.8529 | 0.8533 | 0.8529 | 0.8529 |
88
+ | 0.262 | 28.0 | 1750 | 0.5373 | 0.8563 | 0.8536 | 0.8563 | 0.8543 |
89
+ | 0.232 | 28.99 | 1812 | 0.5384 | 0.8529 | 0.8509 | 0.8529 | 0.8516 |
90
+ | 0.2278 | 29.76 | 1860 | 0.5429 | 0.8512 | 0.8489 | 0.8512 | 0.8495 |
91
 
92
 
93
  ### Framework versions
94
 
95
+ - Transformers 4.37.2
96
+ - Pytorch 2.3.0
97
  - Datasets 2.19.1
98
+ - Tokenizers 0.15.1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:80ccb6002cc52d2f1ca7d86347db3fb7852ea596cb36373fc0159693955fc229
3
  size 88282440
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4dfa44e2974ef6236c8c290b28ce632806f39d090706f142d19229a6ccbb23f6
3
  size 88282440