detr-resnet-50_finetuned_cppe5_v3

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

• Loss: 1.9705
• Map: 0.0365
• Map 50: 0.0659
• Map 75: 0.0314
• Map Small: 0.0246
• Map Medium: 0.0509
• Map Large: 0.0322
• Mar 1: 0.0971
• Mar 10: 0.1951
• Mar 100: 0.254
• Mar Small: 0.159
• Mar Medium: 0.2148
• Mar Large: 0.2409
• Map Coverall: 0.1365
• Mar 100 Coverall: 0.7359
• Map Face Shield: 0.0
• Mar 100 Face Shield: 0.0
• Map Gloves: 0.0185
• Mar 100 Gloves: 0.2527
• Map Goggles: 0.0
• Mar 100 Goggles: 0.0
• Map Mask: 0.0273
• Mar 100 Mask: 0.2812

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: 5e-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: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large	Map Coverall	Mar 100 Coverall	Map Face Shield	Mar 100 Face Shield	Map Gloves	Mar 100 Gloves	Map Goggles	Mar 100 Goggles	Map Mask	Mar 100 Mask
2.7292	1.0	113	2.1368	0.0256	0.0501	0.0217	0.0132	0.0493	0.0237	0.0661	0.1598	0.2091	0.1032	0.1797	0.2083	0.0946	0.6393	0.0	0.0	0.0091	0.1759	0.0	0.0	0.0244	0.2305
2.1942	2.0	226	1.9705	0.0365	0.0659	0.0314	0.0246	0.0509	0.0322	0.0971	0.1951	0.254	0.159	0.2148	0.2409	0.1365	0.7359	0.0	0.0	0.0185	0.2527	0.0	0.0	0.0273	0.2812

Framework versions

• Transformers 4.42.4
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1