<!-- 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. -->
glpn-nyu-finetuned-diode-221121-113853
This model is a fine-tuned version of vinvino02/glpn-nyu on the diode-subset dataset. It achieves the following results on the evaluation set:
- Loss: 0.3384
- Mae: 0.2739
- Rmse: 0.3959
- Abs Rel: 0.3230
- Log Mae: 0.1148
- Log Rmse: 0.1651
- Delta1: 0.5576
- Delta2: 0.8345
- Delta3: 0.9398
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: 1e-05
- train_batch_size: 24
- eval_batch_size: 48
- seed: 2022
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Mae | Rmse | Abs Rel | Log Mae | Log Rmse | Delta1 | Delta2 | Delta3 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.7523 | 1.0 | 72 | 0.5772 | 0.7466 | 0.9116 | 0.9709 | 0.2568 | 0.3040 | 0.1373 | 0.3324 | 0.6328 |
| 0.4281 | 2.0 | 144 | 0.3849 | 0.3324 | 0.4673 | 0.3934 | 0.1349 | 0.1874 | 0.4681 | 0.7753 | 0.9142 |
| 0.3906 | 3.0 | 216 | 0.3660 | 0.3048 | 0.4418 | 0.3593 | 0.1258 | 0.1800 | 0.5225 | 0.7997 | 0.9195 |
| 0.3766 | 4.0 | 288 | 0.3556 | 0.2923 | 0.4275 | 0.3383 | 0.1209 | 0.1741 | 0.5450 | 0.8157 | 0.9259 |
| 0.3744 | 5.0 | 360 | 0.3539 | 0.2899 | 0.4171 | 0.3435 | 0.1208 | 0.1724 | 0.5355 | 0.8173 | 0.9307 |
| 0.328 | 6.0 | 432 | 0.3498 | 0.2860 | 0.4109 | 0.3418 | 0.1196 | 0.1709 | 0.5402 | 0.8193 | 0.9334 |
| 0.3166 | 7.0 | 504 | 0.3451 | 0.2793 | 0.4110 | 0.3203 | 0.1166 | 0.1677 | 0.5583 | 0.8286 | 0.9331 |
| 0.2639 | 8.0 | 576 | 0.3475 | 0.2823 | 0.4083 | 0.3341 | 0.1182 | 0.1695 | 0.5469 | 0.8251 | 0.9337 |
| 0.2802 | 9.0 | 648 | 0.3422 | 0.2779 | 0.4030 | 0.3249 | 0.1163 | 0.1667 | 0.5524 | 0.8287 | 0.9366 |
| 0.2701 | 10.0 | 720 | 0.3411 | 0.2781 | 0.3962 | 0.3316 | 0.1168 | 0.1664 | 0.5446 | 0.8286 | 0.9396 |
| 0.2232 | 11.0 | 792 | 0.3408 | 0.2755 | 0.3998 | 0.3259 | 0.1154 | 0.1665 | 0.5578 | 0.8332 | 0.9383 |
| 0.2921 | 12.0 | 864 | 0.3391 | 0.2749 | 0.3975 | 0.3220 | 0.1152 | 0.1652 | 0.5553 | 0.8332 | 0.9390 |
| 0.2837 | 13.0 | 936 | 0.3400 | 0.2745 | 0.3979 | 0.3251 | 0.1150 | 0.1660 | 0.5587 | 0.8347 | 0.9386 |
| 0.2922 | 14.0 | 1008 | 0.3370 | 0.2728 | 0.3965 | 0.3184 | 0.1142 | 0.1644 | 0.5602 | 0.8359 | 0.9401 |
| 0.2921 | 15.0 | 1080 | 0.3384 | 0.2739 | 0.3959 | 0.3230 | 0.1148 | 0.1651 | 0.5576 | 0.8345 | 0.9398 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu116
- Tokenizers 0.13.2
