detr-r50-cd45rb-all-16ah

This model is a fine-tuned version of facebook/detr-resnet-50 on the cd45rb dataset. It achieves the following results on the evaluation set:

• Loss: 1.6073

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
2.3982	1.0	2303	1.8529
2.1913	2.0	4606	1.7870
2.1311	3.0	6909	1.7361
2.1099	4.0	9212	1.7355
2.0772	5.0	11515	1.7197
2.0559	6.0	13818	1.6910
2.0406	7.0	16121	1.7068
2.0332	8.0	18424	1.6706
2.0238	9.0	20727	1.6655
2.0049	10.0	23030	1.6604
1.9879	11.0	25333	1.6595
1.9898	12.0	27636	1.6804
1.9818	13.0	29939	1.6361
1.9731	14.0	32242	1.6457
1.9584	15.0	34545	1.6221
1.9536	16.0	36848	1.6137
1.9396	17.0	39151	1.6114
1.9332	18.0	41454	1.6085
1.9205	19.0	43757	1.6076
1.924	20.0	46060	1.6073

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1
• Datasets 2.12.0
• Tokenizers 0.13.3