license-plate-detr-dinov3

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.4069

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 8
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
5.9639	0.3408	500	5.7109
3.7311	0.6817	1000	5.8289
2.9953	1.0225	1500	5.5977
2.4684	1.3633	2000	4.9900
2.1458	1.7042	2500	3.0389
1.8966	2.0450	3000	2.9854
1.711	2.3858	3500	2.8960
1.7188	2.7267	4000	2.8657
1.6345	3.0675	4500	2.9198
1.6162	3.4083	5000	3.0553
1.5216	3.7491	5500	3.0822
1.4746	4.0900	6000	3.2468
1.4821	4.4308	6500	3.0554
1.4824	4.7716	7000	2.6149
1.4697	5.1125	7500	2.5948
1.3859	5.4533	8000	2.5043
1.4207	5.7941	8500	2.7286
1.3432	6.1350	9000	2.6156
1.3261	6.4758	9500	2.4771
1.3504	6.8166	10000	2.7247
1.2941	7.1575	10500	2.4605
1.3013	7.4983	11000	2.4045
1.3526	7.8391	11500	2.4069

Framework versions

• Transformers 4.57.0.dev0
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1