detr-r50-cd45rb-all-4ah

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: 1.5882

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
2.4171	1.0	2303	1.8410
2.1862	2.0	4606	1.7627
2.1252	3.0	6909	1.7529
2.1065	4.0	9212	1.7416
2.0764	5.0	11515	1.6991
2.051	6.0	13818	1.6863
2.039	7.0	16121	1.6824
2.0266	8.0	18424	1.6706
2.0177	9.0	20727	1.6724
2.006	10.0	23030	1.6439
1.9853	11.0	25333	1.6503
1.9794	12.0	27636	1.6337
1.9676	13.0	29939	1.6213
1.9646	14.0	32242	1.6213
1.9551	15.0	34545	1.6144
1.9448	16.0	36848	1.6081
1.9347	17.0	39151	1.5986
1.9287	18.0	41454	1.5924
1.9177	19.0	43757	1.5907
1.9216	20.0	46060	1.5882

Framework versions

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