detr-resnet-50-dc5-fashionpedia-finetuned

This model is a fine-tuned version of facebook/detr-resnet-50-dc5 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.2904

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
4.7281	0.0438	50	4.3705
3.9353	0.0876	100	4.0499
4.5369	0.1315	150	3.8890
3.9156	0.1753	200	3.7630
3.6006	0.2191	250	3.6861
3.6562	0.2629	300	3.6110
3.7636	0.3067	350	3.5906
4.0293	0.3506	400	3.5405
3.533	0.3944	450	3.4906
3.1302	0.4382	500	3.4249
3.8257	0.4820	550	3.3910
2.9622	0.5259	600	3.3622
3.9213	0.5697	650	3.3310
4.4062	0.6135	700	3.3303
4.3076	0.6573	750	3.3105
4.0868	0.7011	800	3.3040
4.0639	0.7450	850	3.3076
4.7454	0.7888	900	3.2996
4.3044	0.8326	950	3.2935
3.9519	0.8764	1000	3.2904

Framework versions

• Transformers 4.51.0
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1