<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
vit-large-binary-isic-sharpened-patch-32
This model is a fine-tuned version of google/vit-large-patch32-224-in21k on the ahishamm/isic_binary_sharpened dataset. It achieves the following results on the evaluation set:
- Loss: 0.2092
- Accuracy: 0.9202
- Recall: 0.9202
- F1: 0.9202
- Precision: 0.9202
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Recall | F1 | Precision |
---|---|---|---|---|---|---|---|
0.3437 | 0.09 | 100 | 0.3367 | 0.8412 | 0.8412 | 0.8412 | 0.8412 |
0.3702 | 0.18 | 200 | 0.3094 | 0.8585 | 0.8585 | 0.8585 | 0.8585 |
0.2693 | 0.28 | 300 | 0.4361 | 0.8007 | 0.8007 | 0.8007 | 0.8007 |
0.3183 | 0.37 | 400 | 0.2955 | 0.8643 | 0.8643 | 0.8643 | 0.8643 |
0.2688 | 0.46 | 500 | 0.3064 | 0.8603 | 0.8603 | 0.8603 | 0.8603 |
0.2507 | 0.55 | 600 | 0.3556 | 0.8329 | 0.8329 | 0.8329 | 0.8329 |
0.203 | 0.65 | 700 | 0.3134 | 0.8433 | 0.8433 | 0.8433 | 0.8433 |
0.2315 | 0.74 | 800 | 0.2525 | 0.8856 | 0.8856 | 0.8856 | 0.8856 |
0.3527 | 0.83 | 900 | 0.2815 | 0.8731 | 0.8731 | 0.8731 | 0.8731 |
0.292 | 0.92 | 1000 | 0.3879 | 0.8534 | 0.8534 | 0.8534 | 0.8534 |
0.1342 | 1.02 | 1100 | 0.2927 | 0.8874 | 0.8874 | 0.8874 | 0.8874 |
0.1571 | 1.11 | 1200 | 0.2560 | 0.8912 | 0.8912 | 0.8912 | 0.8912 |
0.1787 | 1.2 | 1300 | 0.3245 | 0.8789 | 0.8789 | 0.8789 | 0.8789 |
0.1757 | 1.29 | 1400 | 0.3308 | 0.8720 | 0.8720 | 0.8720 | 0.8720 |
0.1867 | 1.39 | 1500 | 0.2716 | 0.8876 | 0.8876 | 0.8876 | 0.8876 |
0.124 | 1.48 | 1600 | 0.3663 | 0.8744 | 0.8744 | 0.8744 | 0.8744 |
0.082 | 1.57 | 1700 | 0.2793 | 0.9034 | 0.9034 | 0.9034 | 0.9034 |
0.1365 | 1.66 | 1800 | 0.2399 | 0.9077 | 0.9077 | 0.9077 | 0.9077 |
0.0998 | 1.76 | 1900 | 0.3361 | 0.8901 | 0.8901 | 0.8901 | 0.8901 |
0.0748 | 1.85 | 2000 | 0.3239 | 0.8960 | 0.8960 | 0.8960 | 0.8960 |
0.1163 | 1.94 | 2100 | 0.2092 | 0.9202 | 0.9202 | 0.9202 | 0.9202 |
0.0604 | 2.03 | 2200 | 0.3056 | 0.9139 | 0.9139 | 0.9139 | 0.9139 |
0.0792 | 2.13 | 2300 | 0.2880 | 0.9071 | 0.9071 | 0.9071 | 0.9071 |
0.0749 | 2.22 | 2400 | 0.3015 | 0.9070 | 0.9070 | 0.9070 | 0.9070 |
0.0032 | 2.31 | 2500 | 0.3685 | 0.9090 | 0.9090 | 0.9090 | 0.9090 |
0.1038 | 2.4 | 2600 | 0.3539 | 0.9075 | 0.9075 | 0.9075 | 0.9075 |
0.0474 | 2.5 | 2700 | 0.3220 | 0.9152 | 0.9152 | 0.9152 | 0.9152 |
0.0376 | 2.59 | 2800 | 0.2926 | 0.9203 | 0.9203 | 0.9203 | 0.9203 |
0.0424 | 2.68 | 2900 | 0.3463 | 0.9065 | 0.9065 | 0.9065 | 0.9065 |
0.0408 | 2.77 | 3000 | 0.2772 | 0.9263 | 0.9263 | 0.9263 | 0.9263 |
0.0467 | 2.87 | 3100 | 0.2963 | 0.9227 | 0.9227 | 0.9227 | 0.9227 |
0.0083 | 2.96 | 3200 | 0.2971 | 0.9203 | 0.9203 | 0.9203 | 0.9203 |
0.0165 | 3.05 | 3300 | 0.3162 | 0.9257 | 0.9257 | 0.9257 | 0.9257 |
0.0023 | 3.14 | 3400 | 0.3147 | 0.9267 | 0.9267 | 0.9267 | 0.9267 |
0.0009 | 3.23 | 3500 | 0.3433 | 0.9266 | 0.9266 | 0.9266 | 0.9266 |
0.0007 | 3.33 | 3600 | 0.3216 | 0.9312 | 0.9312 | 0.9312 | 0.9312 |
0.0011 | 3.42 | 3700 | 0.3209 | 0.9346 | 0.9346 | 0.9346 | 0.9346 |
0.0029 | 3.51 | 3800 | 0.3236 | 0.9325 | 0.9325 | 0.9325 | 0.9325 |
0.0011 | 3.6 | 3900 | 0.3297 | 0.9302 | 0.9302 | 0.9302 | 0.9302 |
0.0225 | 3.7 | 4000 | 0.3263 | 0.9323 | 0.9323 | 0.9323 | 0.9323 |
0.0008 | 3.79 | 4100 | 0.3352 | 0.9311 | 0.9311 | 0.9311 | 0.9311 |
0.0391 | 3.88 | 4200 | 0.3343 | 0.9282 | 0.9282 | 0.9282 | 0.9282 |
0.0019 | 3.97 | 4300 | 0.3319 | 0.9280 | 0.9280 | 0.9280 | 0.9280 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3