detr-cppe5-ppe-detector

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: 1.1687

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 with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	106	1.9109
No log	2.0	212	1.7408
No log	3.0	318	1.7527
No log	4.0	424	1.5295
3.5859	5.0	530	1.5355
3.5859	6.0	636	1.5048
3.5859	7.0	742	1.4679
3.5859	8.0	848	1.4037
3.5859	9.0	954	1.4161
1.3248	10.0	1060	1.3470
1.3248	11.0	1166	1.3197
1.3248	12.0	1272	1.3047
1.3248	13.0	1378	1.3000
1.3248	14.0	1484	1.2670
1.1606	15.0	1590	1.2603
1.1606	16.0	1696	1.2669
1.1606	17.0	1802	1.2255
1.1606	18.0	1908	1.2329
1.0363	19.0	2014	1.1964
1.0363	20.0	2120	1.1991
1.0363	21.0	2226	1.1966
1.0363	22.0	2332	1.1834
1.0363	23.0	2438	1.1747
0.938	24.0	2544	1.1785
0.938	25.0	2650	1.1752
0.938	26.0	2756	1.1718
0.938	27.0	2862	1.1747
0.938	28.0	2968	1.1736
0.8943	29.0	3074	1.1734
0.8943	30.0	3180	1.1687

Framework versions

• Transformers 4.49.0
• Pytorch 2.11.0+cu128
• Datasets 2.21.0
• Tokenizers 0.21.4