1.9 KiB
1.9 KiB
| license | tags | model-index | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| apache-2.0 |
|
|
glpn-kitti-finetuned-diode
This model is a fine-tuned version of vinvino02/glpn-kitti on the diode-subset dataset. It achieves the following results on the evaluation set:
- Loss: 0.4947
- Rmse: 0.4399
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Rmse |
|---|---|---|---|---|
| 0.8555 | 1.0 | 10 | 0.6764 | 0.5844 |
| 0.7688 | 2.0 | 20 | 0.6335 | 0.5344 |
| 0.7099 | 3.0 | 30 | 0.5967 | 0.5203 |
| 0.6526 | 4.0 | 40 | 0.5602 | 0.4932 |
| 0.5928 | 5.0 | 50 | 0.5235 | 0.4576 |
| 0.5746 | 6.0 | 60 | 0.4981 | 0.4160 |
| 0.5335 | 7.0 | 70 | 0.4794 | 0.3926 |
| 0.5009 | 8.0 | 80 | 0.4697 | 0.3816 |
| 0.4811 | 9.0 | 90 | 0.4629 | nan |
| 0.473 | 10.0 | 100 | 0.4633 | 0.3714 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.2