detr-resnet-50_sgd_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: 7.6389

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: 0.0001
• 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
7.2801	0.77	50	7.6418
7.3084	1.54	100	7.6658
7.3307	2.31	150	7.5842
7.287	3.08	200	7.6073
7.3372	3.85	250	7.7152
7.3168	4.62	300	7.6034
7.3691	5.38	350	7.6833
7.2893	6.15	400	7.6372
7.3304	6.92	450	7.6150
7.2394	7.69	500	7.6044
7.4489	8.46	550	7.6391
7.1353	9.23	600	7.7143
7.3786	10.0	650	7.6010
7.289	10.77	700	7.6191
7.2943	11.54	750	7.6662
7.2791	12.31	800	7.6043
7.2665	13.08	850	7.6339
7.2157	13.85	900	7.6102
7.309	14.62	950	7.6389

Framework versions

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