detr_finetuned_trashify_box_detector_with_data_aug

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

• Loss: 1.0704

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 25
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
100.4735	1.0	50	8.0297
4.369	2.0	100	2.7376
2.5518	3.0	150	2.1839
2.2226	4.0	200	1.9228
1.9906	5.0	250	1.7408
1.8219	6.0	300	1.5573
1.6974	7.0	350	1.4779
1.6027	8.0	400	1.4510
1.5517	9.0	450	1.3711
1.4491	10.0	500	1.3177
1.4335	11.0	550	1.2811
1.3645	12.0	600	1.2475
1.3314	13.0	650	1.2060
1.2973	14.0	700	1.1874
1.2506	15.0	750	1.1794
1.2319	16.0	800	1.1657
1.1479	17.0	850	1.1300
1.1466	18.0	900	1.1179
1.1138	19.0	950	1.1095
1.1153	20.0	1000	1.0961
1.0894	21.0	1050	1.0790
1.0691	22.0	1100	1.0870
1.0619	23.0	1150	1.0804
1.0459	24.0	1200	1.0717
1.0363	25.0	1250	1.0704

Framework versions

• Transformers 4.45.0.dev0
• Pytorch 2.4.0+cu124
• Datasets 2.21.0
• Tokenizers 0.19.1