Hugging Face's logo

lombardata
/
drone-DinoVdeau-produttoria-probabilities-large-2024_11_04-batch-size64_freeze_probs

TensorBoard
Safetensors
English
dinov2
multilabel-image-classification
multilabel
Generated from Trainer
Model card Files Metrics Community
drone-DinoVdeau-produttoria-probabilities-large-2024_11_04-batch-size64_freeze_probs / README.md
lombardata's picture
lombardata
Evaluation on the test set completed on 2024_11_05.
2ae096e verified
preview code
|
raw
history blame
9.15 kB
metadata
license: apache-2.0
base_model: facebook/dinov2-large
tags:
  - generated_from_trainer
model-index:
  - name: >-
      drone-DinoVdeau-produttoria-probabilities-large-2024_11_04-batch-size64_freeze_probs
    results: []

drone-DinoVdeau-produttoria-probabilities-large-2024_11_04-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: 0.3194
  • Rmse: 0.2405
  • Mae: 0.1536
  • R2: 0.4281
  • Explained Variance: 0.4294
  • 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: 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 0.3610 0.2645 0.1878 0.2818 0.3037 0.001
No log 2.0 362 0.3465 0.2566 0.1778 0.3349 0.3479 0.001
0.4149 3.0 543 0.3413 0.2532 0.1731 0.3536 0.3600 0.001
0.4149 4.0 724 0.3406 0.2532 0.1743 0.3519 0.3591 0.001
0.4149 5.0 905 0.3342 0.2496 0.1661 0.3702 0.3731 0.001
0.3486 6.0 1086 0.3385 0.2512 0.1739 0.3651 0.3724 0.001
0.3486 7.0 1267 0.3321 0.2476 0.1650 0.3836 0.3846 0.001
0.3486 8.0 1448 0.3332 0.2484 0.1629 0.3802 0.3811 0.001
0.3462 9.0 1629 0.3305 0.2468 0.1652 0.3859 0.3872 0.001
0.3462 10.0 1810 0.3314 0.2476 0.1655 0.3827 0.3850 0.001
0.3462 11.0 1991 0.3320 0.2474 0.1602 0.3840 0.3866 0.001
0.3391 12.0 2172 0.3342 0.2494 0.1683 0.3761 0.3843 0.001
0.3391 13.0 2353 0.3325 0.2480 0.1649 0.3821 0.3836 0.001
0.3372 14.0 2534 0.3323 0.2472 0.1700 0.3878 0.3930 0.001
0.3372 15.0 2715 0.3349 0.2493 0.1703 0.3749 0.3828 0.001
0.3372 16.0 2896 0.3279 0.2448 0.1649 0.3983 0.4019 0.0001
0.3343 17.0 3077 0.3279 0.2448 0.1648 0.3984 0.4041 0.0001
0.3343 18.0 3258 0.3262 0.2440 0.1622 0.4025 0.4032 0.0001
0.3343 19.0 3439 0.3247 0.2432 0.1588 0.4046 0.4051 0.0001
0.3271 20.0 3620 0.3261 0.2433 0.1625 0.4059 0.4106 0.0001
0.3271 21.0 3801 0.3241 0.2424 0.1606 0.4095 0.4119 0.0001
0.3271 22.0 3982 0.3236 0.2422 0.1587 0.4111 0.4132 0.0001
0.3275 23.0 4163 0.3242 0.2423 0.1601 0.4107 0.4134 0.0001
0.3275 24.0 4344 0.3227 0.2414 0.1586 0.4150 0.4161 0.0001
0.3247 25.0 4525 0.3224 0.2413 0.1587 0.4148 0.4162 0.0001
0.3247 26.0 4706 0.3218 0.2413 0.1557 0.4143 0.4155 0.0001
0.3247 27.0 4887 0.3227 0.2416 0.1603 0.4138 0.4154 0.0001
0.3231 28.0 5068 0.3207 0.2405 0.1562 0.4186 0.4197 0.0001
0.3231 29.0 5249 0.3221 0.2411 0.1597 0.4163 0.4175 0.0001
0.3231 30.0 5430 0.3225 0.2413 0.1608 0.4164 0.4190 0.0001
0.3215 31.0 5611 0.3224 0.2416 0.1535 0.4134 0.4164 0.0001
0.3215 32.0 5792 0.3213 0.2408 0.1553 0.4180 0.4185 0.0001
0.3215 33.0 5973 0.3216 0.2414 0.1583 0.4123 0.4142 0.0001
0.3227 34.0 6154 0.3205 0.2406 0.1562 0.4172 0.4181 0.0001
0.3227 35.0 6335 0.3198 0.2399 0.1535 0.4215 0.4224 0.0001
0.3202 36.0 6516 0.3211 0.2406 0.1577 0.4187 0.4194 0.0001
0.3202 37.0 6697 0.3204 0.2403 0.1520 0.4188 0.4203 0.0001
0.3202 38.0 6878 0.3214 0.2409 0.1560 0.4170 0.4185 0.0001
0.3195 39.0 7059 0.3195 0.2397 0.1520 0.4226 0.4232 0.0001
0.3195 40.0 7240 0.3208 0.2404 0.1577 0.4204 0.4231 0.0001
0.3195 41.0 7421 0.3198 0.2398 0.1547 0.4217 0.4233 0.0001
0.3192 42.0 7602 0.3218 0.2410 0.1589 0.4174 0.4218 0.0001
0.3192 43.0 7783 0.3190 0.2396 0.1544 0.4235 0.4254 0.0001
0.3192 44.0 7964 0.3190 0.2396 0.1534 0.4230 0.4239 0.0001
0.3178 45.0 8145 0.3198 0.2397 0.1566 0.4239 0.4260 0.0001
0.3178 46.0 8326 0.3193 0.2398 0.1556 0.4213 0.4231 0.0001
0.3175 47.0 8507 0.3190 0.2393 0.1524 0.4245 0.4257 0.0001
0.3175 48.0 8688 0.3193 0.2398 0.1525 0.4215 0.4230 0.0001
0.3175 49.0 8869 0.3207 0.2405 0.1558 0.4187 0.4196 0.0001
0.3174 50.0 9050 0.3198 0.2400 0.1572 0.4218 0.4237 1e-05
0.3174 51.0 9231 0.3244 0.2426 0.1602 0.4092 0.4173 1e-05
0.3174 52.0 9412 0.3190 0.2396 0.1550 0.4227 0.4235 1e-05
0.3152 53.0 9593 0.3189 0.2394 0.1552 0.4249 0.4270 1e-05
0.3152 54.0 9774 0.3194 0.2396 0.1540 0.4227 0.4239 1e-05
0.3152 55.0 9955 0.3185 0.2391 0.1539 0.4250 0.4258 1e-05
0.317 56.0 10136 0.3181 0.2388 0.1527 0.4273 0.4281 1e-05
0.317 57.0 10317 0.3187 0.2392 0.1532 0.4259 0.4274 1e-05
0.317 58.0 10498 0.3201 0.2401 0.1567 0.4217 0.4259 1e-05
0.314 59.0 10679 0.3181 0.2388 0.1528 0.4270 0.4282 1e-05
0.314 60.0 10860 0.3182 0.2389 0.1534 0.4256 0.4268 1e-05
0.314 61.0 11041 0.3186 0.2391 0.1510 0.4255 0.4266 1e-05
0.314 62.0 11222 0.3203 0.2398 0.1596 0.4240 0.4262 1e-05
0.314 63.0 11403 0.3196 0.2397 0.1570 0.4242 0.4276 1e-05
0.3142 64.0 11584 0.3181 0.2391 0.1527 0.4244 0.4253 1e-05
0.3142 65.0 11765 0.3185 0.2390 0.1550 0.4259 0.4264 1e-05
0.3142 66.0 11946 0.3186 0.2389 0.1562 0.4278 0.4291 0.0000
0.3131 67.0 12127 0.3181 0.2387 0.1526 0.4270 0.4279 0.0000
0.3131 68.0 12308 0.3195 0.2397 0.1549 0.4221 0.4257 0.0000
0.3131 69.0 12489 0.3183 0.2390 0.1540 0.4259 0.4275 0.0000

Framework versions

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