detr-r101-cd45rb-all-4ah

This model is a fine-tuned version of facebook/detr-resnet-101 on the cd45rb dataset. It achieves the following results on the evaluation set:

• Loss: 1.5962

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
2.3757	1.0	2303	1.8222
2.1779	2.0	4606	1.8000
2.1259	3.0	6909	1.7306
2.0969	4.0	9212	1.6854
2.0699	5.0	11515	1.6934
2.0398	6.0	13818	1.6655
2.0203	7.0	16121	1.6576
2.009	8.0	18424	1.6401
1.9923	9.0	20727	1.6319
1.9802	10.0	23030	1.6264
2.0027	11.0	25333	1.6539
2.0005	12.0	27636	1.6564
1.9861	13.0	29939	1.6383
1.9778	14.0	32242	1.6246
1.9602	15.0	34545	1.6119
1.9599	16.0	36848	1.6097
1.9464	17.0	39151	1.6115
1.9376	18.0	41454	1.6018
1.9273	19.0	43757	1.5967
1.928	20.0	46060	1.5962

Framework versions

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