Hugging Face's logo

lombardata
/
drone-DinoVdeau-produttoria_binary-probabilities-large-2024_11_03-batch-size64_freeze_probs

TensorBoard
Safetensors
English
dinov2
multilabel-image-classification
multilabel
Generated from Trainer
Model card Files Metrics Community
drone-DinoVdeau-produttoria_binary-probabilities-large-2024_11_03-batch-size64_freeze_probs / README.md
lombardata's picture
lombardata
Evaluation on the test set completed on 2024_11_04.
2db9804 verified
preview code
|
raw
history blame
9.06 kB
metadata
license: apache-2.0
base_model: facebook/dinov2-large
tags:
  - generated_from_trainer
model-index:
  - name: >-
      drone-DinoVdeau-produttoria_binary-probabilities-large-2024_11_03-batch-size64_freeze_probs
    results: []

drone-DinoVdeau-produttoria_binary-probabilities-large-2024_11_03-batch-size64_freeze_probs

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

  • Loss: 0.3499
  • Rmse: 0.1848
  • Mae: 0.1248
  • R2: 0.4361
  • Explained Variance: 0.4376
  • Learning Rate: 1e-05

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: 64
  • eval_batch_size: 64
  • 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 Rmse Mae R2 Explained Variance Rate
No log 1.0 181 0.3795 0.2067 0.1489 0.2894 0.3009 0.001
No log 2.0 362 0.3674 0.1983 0.1374 0.3517 0.3548 0.001
0.4416 3.0 543 0.3671 0.1981 0.1414 0.3521 0.3569 0.001
0.4416 4.0 724 0.3632 0.1952 0.1391 0.3708 0.3749 0.001
0.4416 5.0 905 0.3679 0.1993 0.1418 0.3453 0.3614 0.001
0.3813 6.0 1086 0.3625 0.1951 0.1380 0.3718 0.3743 0.001
0.3813 7.0 1267 0.3619 0.1941 0.1348 0.3771 0.3837 0.001
0.3813 8.0 1448 0.3613 0.1935 0.1368 0.3788 0.3809 0.001
0.3785 9.0 1629 0.3604 0.1934 0.1354 0.3812 0.3833 0.001
0.3785 10.0 1810 0.3613 0.1932 0.1338 0.3812 0.3844 0.001
0.3785 11.0 1991 0.3604 0.1931 0.1323 0.3845 0.3857 0.001
0.3743 12.0 2172 0.3618 0.1942 0.1386 0.3774 0.3844 0.001
0.3743 13.0 2353 0.3593 0.1925 0.1343 0.3875 0.3894 0.001
0.3732 14.0 2534 0.3605 0.1932 0.1352 0.3831 0.3863 0.001
0.3732 15.0 2715 0.3605 0.1935 0.1366 0.3817 0.3836 0.001
0.3732 16.0 2896 0.3600 0.1922 0.1312 0.3882 0.3910 0.001
0.3733 17.0 3077 0.3629 0.1932 0.1378 0.3843 0.3882 0.001
0.3733 18.0 3258 0.3615 0.1943 0.1323 0.3768 0.3840 0.001
0.3733 19.0 3439 0.3595 0.1922 0.1330 0.3895 0.3911 0.001
0.3723 20.0 3620 0.3566 0.1902 0.1330 0.4006 0.4041 0.0001
0.3723 21.0 3801 0.3549 0.1890 0.1306 0.4076 0.4089 0.0001
0.3723 22.0 3982 0.3545 0.1886 0.1308 0.4096 0.4108 0.0001
0.3683 23.0 4163 0.3545 0.1882 0.1303 0.4116 0.4124 0.0001
0.3683 24.0 4344 0.3540 0.1882 0.1317 0.4121 0.4131 0.0001
0.3654 25.0 4525 0.3546 0.1883 0.1284 0.4113 0.4126 0.0001
0.3654 26.0 4706 0.3529 0.1876 0.1264 0.4154 0.4165 0.0001
0.3654 27.0 4887 0.3533 0.1874 0.1294 0.4166 0.4177 0.0001
0.3652 28.0 5068 0.3532 0.1876 0.1294 0.4160 0.4169 0.0001
0.3652 29.0 5249 0.3531 0.1871 0.1302 0.4184 0.4192 0.0001
0.3652 30.0 5430 0.3536 0.1878 0.1292 0.4148 0.4160 0.0001
0.3628 31.0 5611 0.3531 0.1877 0.1267 0.4152 0.4175 0.0001
0.3628 32.0 5792 0.3528 0.1876 0.1288 0.4162 0.4168 0.0001
0.3628 33.0 5973 0.3515 0.1864 0.1273 0.4225 0.4230 0.0001
0.3638 34.0 6154 0.3520 0.1868 0.1263 0.4202 0.4216 0.0001
0.3638 35.0 6335 0.3518 0.1866 0.1278 0.4215 0.4220 0.0001
0.3618 36.0 6516 0.3523 0.1871 0.1285 0.4193 0.4196 0.0001
0.3618 37.0 6697 0.3516 0.1866 0.1273 0.4217 0.4225 0.0001
0.3618 38.0 6878 0.3527 0.1878 0.1274 0.4157 0.4184 0.0001
0.3611 39.0 7059 0.3512 0.1862 0.1266 0.4242 0.4249 0.0001
0.3611 40.0 7240 0.3521 0.1866 0.1302 0.4224 0.4237 0.0001
0.3611 41.0 7421 0.3507 0.1858 0.1266 0.4264 0.4275 0.0001
0.3613 42.0 7602 0.3513 0.1860 0.1278 0.4263 0.4272 0.0001
0.3613 43.0 7783 0.3511 0.1860 0.1274 0.4262 0.4273 0.0001
0.3613 44.0 7964 0.3514 0.1859 0.1244 0.4266 0.4282 0.0001
0.3603 45.0 8145 0.3525 0.1863 0.1273 0.4249 0.4276 0.0001
0.3603 46.0 8326 0.3505 0.1856 0.1258 0.4275 0.4286 0.0001
0.3603 47.0 8507 0.3517 0.1866 0.1250 0.4231 0.4258 0.0001
0.3603 48.0 8688 0.3504 0.1856 0.1259 0.4286 0.4292 0.0001
0.3603 49.0 8869 0.3507 0.1857 0.1272 0.4274 0.4284 0.0001
0.3604 50.0 9050 0.3516 0.1857 0.1283 0.4280 0.4289 0.0001
0.3604 51.0 9231 0.3529 0.1867 0.1288 0.4227 0.4282 0.0001
0.3604 52.0 9412 0.3506 0.1857 0.1268 0.4282 0.4295 0.0001
0.3592 53.0 9593 0.3505 0.1856 0.1273 0.4286 0.4302 0.0001
0.3592 54.0 9774 0.3502 0.1854 0.1266 0.4300 0.4304 0.0001
0.3592 55.0 9955 0.3501 0.1854 0.1251 0.4299 0.4319 0.0001
0.3601 56.0 10136 0.3507 0.1858 0.1243 0.4273 0.4294 0.0001
0.3601 57.0 10317 0.3509 0.1860 0.1253 0.4274 0.4297 0.0001
0.3601 58.0 10498 0.3493 0.1846 0.1251 0.4338 0.4354 0.0001
0.3601 59.0 10679 0.3501 0.1855 0.1241 0.4282 0.4299 0.0001
0.3601 60.0 10860 0.3501 0.1852 0.1259 0.4303 0.4325 0.0001
0.3588 61.0 11041 0.3498 0.1850 0.1264 0.4305 0.4310 0.0001
0.3588 62.0 11222 0.3498 0.1850 0.1265 0.4323 0.4333 0.0001
0.3588 63.0 11403 0.3502 0.1851 0.1270 0.4321 0.4339 0.0001
0.3579 64.0 11584 0.3500 0.1853 0.1256 0.4300 0.4312 0.0001
0.3579 65.0 11765 0.3501 0.1854 0.1280 0.4299 0.4304 1e-05
0.3579 66.0 11946 0.3493 0.1847 0.1253 0.4336 0.4342 1e-05
0.3564 67.0 12127 0.3494 0.1847 0.1261 0.4334 0.4340 1e-05
0.3564 68.0 12308 0.3500 0.1856 0.1261 0.4291 0.4307 1e-05

Framework versions

  • Transformers 4.41.0
  • Pytorch 2.5.0+cu124
  • Datasets 3.0.2
  • Tokenizers 0.19.1