detr-resnet-101_adamw_torch_finetuned_food-roboflow

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

• Loss: 3.0549

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: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
6.854	0.77	50	6.6564
5.8031	1.54	100	5.7440
5.0476	2.31	150	4.8884
4.1629	3.08	200	3.9522
3.4843	3.85	250	3.5612
3.0905	4.62	300	3.7328
3.0965	5.38	350	3.3028
2.8764	6.15	400	3.1671
2.8236	6.92	450	3.2082
2.8467	7.69	500	3.0500
2.6769	8.46	550	3.0538
2.7194	9.23	600	3.0982
2.6311	10.0	650	3.0520
2.6772	10.77	700	2.9950
2.577	11.54	750	3.0134
2.6274	12.31	800	3.0523
2.597	13.08	850	3.0120
2.56	13.85	900	2.9795
2.5803	14.62	950	3.0549

Framework versions

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