detr-r50-cd45rb-8ah-6l

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: 1.5794

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: 4
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
2.4794	1.0	4606	1.8245
2.2469	2.0	9212	1.7593
2.1817	3.0	13818	1.7109
2.1418	4.0	18424	1.6976
2.118	5.0	23030	1.6695
2.1237	6.0	27636	1.6781
2.1025	7.0	32242	1.6574
2.0796	8.0	36848	1.6418
2.0672	9.0	41454	1.6333
2.0597	10.0	46060	1.6313
2.0948	11.0	50666	1.6546
2.0943	12.0	55272	1.6905
2.0819	13.0	59878	1.6430
2.0795	14.0	64484	1.6439
2.0566	15.0	69090	1.6449
2.0435	16.0	73696	1.6204
2.0375	17.0	78302	1.6195
2.032	18.0	82908	1.6128
2.0079	19.0	87514	1.6082
1.9985	20.0	92120	1.6037
1.9976	21.0	96726	1.6005
1.9887	22.0	101332	1.5969
1.9841	23.0	105938	1.5841
1.9763	24.0	110544	1.5826
1.9686	25.0	115150	1.5794

Framework versions

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