Hugging Face's logo

lombardata
/
dinov2-base-2024_09_09-batch-size32_epochs150_freeze

TensorBoard
Safetensors
English
dinov2
multilabel-image-classification
multilabel
Generated from Trainer
Model card Files Metrics Community
dinov2-base-2024_09_09-batch-size32_epochs150_freeze / README.md
lombardata's picture
lombardata
Evaluation on the test set completed on 2024_09_10.
b7619ec verified
preview code
|
raw
history blame
7.71 kB
metadata
license: apache-2.0
base_model: facebook/dinov2-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: dinov2-base-2024_09_09-batch-size32_epochs150_freeze
    results: []

dinov2-base-2024_09_09-batch-size32_epochs150_freeze

This model is a fine-tuned version of facebook/dinov2-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1321
  • F1 Micro: 0.8069
  • F1 Macro: 0.7121
  • Roc Auc: 0.8742
  • Accuracy: 0.2869
  • Learning Rate: 0.0000

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: 0.001
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 150
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss F1 Micro F1 Macro Roc Auc Accuracy Rate
No log 1.0 273 0.1601 0.7634 0.6251 0.8453 0.2328 0.001
0.1759 2.0 546 0.1504 0.7780 0.6462 0.8546 0.2498 0.001
0.1759 3.0 819 0.1483 0.7817 0.6644 0.8583 0.2564 0.001
0.1474 4.0 1092 0.1464 0.7863 0.6809 0.8634 0.2554 0.001
0.1474 5.0 1365 0.1423 0.7891 0.6919 0.8572 0.2682 0.001
0.1397 6.0 1638 0.1440 0.7902 0.6988 0.8629 0.2651 0.001
0.1397 7.0 1911 0.1425 0.7938 0.6850 0.8647 0.2682 0.001
0.1356 8.0 2184 0.1429 0.7931 0.6880 0.8700 0.2637 0.001
0.1356 9.0 2457 0.1463 0.7927 0.6885 0.8704 0.2557 0.001
0.1315 10.0 2730 0.1392 0.8009 0.7050 0.8729 0.2744 0.001
0.1308 11.0 3003 0.1443 0.7853 0.6892 0.8519 0.2699 0.001
0.1308 12.0 3276 0.1452 0.7888 0.6976 0.8670 0.2713 0.001
0.1277 13.0 3549 0.1370 0.8007 0.7032 0.8680 0.2765 0.001
0.1277 14.0 3822 0.1401 0.7984 0.6875 0.8694 0.2730 0.001
0.1257 15.0 4095 0.1379 0.8049 0.7001 0.8748 0.2817 0.001
0.1257 16.0 4368 0.1429 0.7969 0.7063 0.8675 0.2682 0.001
0.1257 17.0 4641 0.1451 0.7956 0.6861 0.8728 0.2613 0.001
0.1257 18.0 4914 0.1418 0.7906 0.6849 0.8574 0.2713 0.001
0.1251 19.0 5187 0.1438 0.7900 0.6794 0.8556 0.2654 0.001
0.1251 20.0 5460 0.1319 0.8068 0.7202 0.8705 0.2866 0.0001
0.1161 21.0 5733 0.1312 0.8081 0.7237 0.8715 0.2876 0.0001
0.1109 22.0 6006 0.1310 0.8101 0.7222 0.8788 0.2935 0.0001
0.1109 23.0 6279 0.1305 0.8120 0.7226 0.8776 0.2935 0.0001
0.1103 24.0 6552 0.1309 0.8096 0.7238 0.8769 0.2952 0.0001
0.1103 25.0 6825 0.1308 0.8093 0.7171 0.8735 0.2949 0.0001
0.1099 26.0 7098 0.1301 0.8100 0.7200 0.8745 0.2911 0.0001
0.1099 27.0 7371 0.1303 0.8082 0.7208 0.8727 0.2924 0.0001
0.1107 28.0 7644 0.1302 0.8103 0.7218 0.8752 0.2970 0.0001
0.1107 29.0 7917 0.1302 0.8104 0.7237 0.8766 0.2963 0.0001
0.1103 30.0 8190 0.1303 0.8097 0.7181 0.8745 0.2956 0.0001
0.1103 31.0 8463 0.1301 0.8092 0.7190 0.8739 0.2959 0.0001
0.1104 32.0 8736 0.1301 0.8098 0.7210 0.8740 0.2928 0.0001
0.1093 33.0 9009 0.1296 0.8100 0.7204 0.8738 0.2963 1e-05
0.1093 34.0 9282 0.1296 0.8101 0.7222 0.8745 0.2956 1e-05
0.1084 35.0 9555 0.1295 0.8109 0.7220 0.8758 0.2956 1e-05
0.1084 36.0 9828 0.1295 0.8105 0.7212 0.8746 0.2931 1e-05
0.1091 37.0 10101 0.1295 0.8119 0.7239 0.8757 0.2963 1e-05
0.1091 38.0 10374 0.1295 0.8104 0.7213 0.8744 0.2959 1e-05
0.1075 39.0 10647 0.1295 0.8106 0.7222 0.8752 0.2966 1e-05
0.1075 40.0 10920 0.1295 0.8113 0.7233 0.8768 0.2956 1e-05
0.1088 41.0 11193 0.1295 0.8100 0.7223 0.8739 0.2945 1e-05
0.1088 42.0 11466 0.1295 0.8111 0.7219 0.8750 0.2973 1e-05
0.1085 43.0 11739 0.1294 0.8098 0.7212 0.8738 0.2931 1e-05
0.1084 44.0 12012 0.1295 0.8108 0.7212 0.8746 0.2970 1e-05
0.1084 45.0 12285 0.1294 0.8104 0.7218 0.8749 0.2945 1e-05
0.1083 46.0 12558 0.1294 0.8113 0.7233 0.8759 0.2976 1e-05
0.1083 47.0 12831 0.1294 0.8107 0.7229 0.8753 0.2945 1e-05
0.109 48.0 13104 0.1294 0.8103 0.7209 0.8742 0.2956 1e-05
0.109 49.0 13377 0.1293 0.8111 0.7215 0.8755 0.2959 1e-05
0.108 50.0 13650 0.1294 0.8107 0.7211 0.8750 0.2966 1e-05
0.108 51.0 13923 0.1294 0.8099 0.7224 0.8742 0.2924 1e-05
0.1084 52.0 14196 0.1294 0.8110 0.7224 0.8755 0.2973 1e-05
0.1084 53.0 14469 0.1295 0.8111 0.7225 0.8757 0.2980 1e-05
0.1086 54.0 14742 0.1294 0.8105 0.7222 0.8752 0.2963 1e-05
0.1083 55.0 15015 0.1293 0.8107 0.7231 0.8754 0.2956 1e-05
0.1083 56.0 15288 0.1294 0.8107 0.7227 0.8753 0.2959 0.0000
0.108 57.0 15561 0.1293 0.8111 0.7231 0.8754 0.2956 0.0000
0.108 58.0 15834 0.1294 0.8112 0.7230 0.8755 0.2966 0.0000
0.1089 59.0 16107 0.1294 0.8110 0.7227 0.8753 0.2966 0.0000

Framework versions

  • Transformers 4.41.1
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1