detr_finetuned_roadsigns

This model is a fine-tuned version of microsoft/conditional-detr-resnet-50 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.9152

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: 25

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	35	2.4704
No log	2.0	70	1.6563
No log	3.0	105	1.5779
No log	4.0	140	1.5352
No log	5.0	175	1.4800
No log	6.0	210	1.2647
No log	7.0	245	1.3520
No log	8.0	280	1.2205
No log	9.0	315	1.2303
No log	10.0	350	1.1987
No log	11.0	385	1.0827
No log	12.0	420	1.0480
No log	13.0	455	1.0071
No log	14.0	490	1.0634
3.4885	15.0	525	0.9846
3.4885	16.0	560	0.9775
3.4885	17.0	595	0.9928
3.4885	18.0	630	0.9659
3.4885	19.0	665	0.9488
3.4885	20.0	700	0.9296
3.4885	21.0	735	0.9195
3.4885	22.0	770	0.9125
3.4885	23.0	805	0.9158
3.4885	24.0	840	0.9151
3.4885	25.0	875	0.9152

Framework versions

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