detr_finetuned_roadsigns_comparison

This model is a fine-tuned version of Jesse020202/detr_finetuned_roadsigns on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7791

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• 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: cosine
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	35	1.0460
No log	2.0	70	0.9895
No log	3.0	105	0.9574
No log	4.0	140	0.9357
No log	5.0	175	0.9303
No log	6.0	210	0.9347
No log	7.0	245	1.0464
No log	8.0	280	0.8754
No log	9.0	315	0.8770
No log	10.0	350	0.7999
No log	11.0	385	0.8237
No log	12.0	420	0.8142
No log	13.0	455	0.8347
No log	14.0	490	0.7893
0.9165	15.0	525	0.7876
0.9165	16.0	560	0.7874
0.9165	17.0	595	0.7868
0.9165	18.0	630	0.7794
0.9165	19.0	665	0.7793
0.9165	20.0	700	0.7791

Framework versions

• Transformers 4.56.2
• Pytorch 2.8.0
• Datasets 4.1.1
• Tokenizers 0.22.1