detr-r50-cd45rb-all-8ah

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: 1.6306

• model.num_parameters() 41509318

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
2.4211	1.0	2303	1.8771
2.2126	2.0	4606	1.8058
2.1291	3.0	6909	1.7395
2.1026	4.0	9212	1.7130
2.0756	5.0	11515	1.7131
2.0535	6.0	13818	1.6874
2.0333	7.0	16121	1.6741
2.022	8.0	18424	1.6632
2.007	9.0	20727	1.6500
1.9998	10.0	23030	1.6492
1.9945	11.0	25333	1.6559
1.9906	12.0	27636	1.6375
1.9831	13.0	29939	1.6384
1.9787	14.0	32242	1.6331
1.9686	15.0	34545	1.6298
2.0042	16.0	36848	1.6546
2.0043	17.0	39151	1.6496
1.9945	18.0	41454	1.6571
1.9663	19.0	43757	1.6379
1.9635	20.0	46060	1.6306

Framework versions

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