detr-resnet-50_adamw_torch_fused_adam_beta1-0_8_finetuned_food-roboflow

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

• Loss: 2.8078

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.8,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
6.6897	1.52	50	6.2985
5.5704	3.03	100	5.1916
4.3645	4.55	150	4.0882
3.4115	6.06	200	3.3534
3.0109	7.58	250	3.0666
2.8836	9.09	300	3.0176
2.7564	10.61	350	2.9294
2.7007	12.12	400	2.9228
2.7032	13.64	450	2.8650
2.6248	15.15	500	2.8833
2.5995	16.67	550	2.8085
2.5616	18.18	600	2.7843
2.5582	19.7	650	2.8078

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1