kvasir_seg_rtdetr_r18_test_fps

This model is a fine-tuned version of PekingU/rtdetr_r18vd_coco_o365 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 6.4708
• Map: 0.5375
• Map 50: 0.7037
• Map 75: 0.6008
• Map Small: 0.0
• Map Medium: 0.2762
• Map Large: 0.5563
• Mar 1: 0.6024
• Mar 10: 0.8218
• Mar 100: 0.9209
• Mar Small: 0.0
• Mar Medium: 0.78
• Mar Large: 0.9325
• Map Polyp: 0.5375
• Mar 100 Polyp: 0.9209

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: 1
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 300
• num_epochs: 10

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 Polyp	Mar 100 Polyp
188.8938	1.0	100	35.3914	0.045	0.0921	0.0405	0.0	0.0001	0.0612	0.1175	0.3521	0.6398	0.0	0.09	0.6705	0.045	0.6398
20.5545	2.0	200	13.3860	0.0707	0.1207	0.0633	0.0	0.0164	0.0763	0.2	0.5607	0.8398	0.0	0.63	0.8545	0.0707	0.8398
12.8462	3.0	300	8.8009	0.251	0.3922	0.2349	0.0	0.1155	0.263	0.3758	0.7171	0.8782	0.0	0.71	0.891	0.251	0.8782
11.3	4.0	400	8.9424	0.2968	0.4137	0.3193	0.0	0.2263	0.3046	0.3929	0.7687	0.9038	0.0	0.77	0.915	0.2968	0.9038
10.3262	5.0	500	7.8881	0.2819	0.4434	0.3041	0.0	0.2255	0.2897	0.4223	0.7374	0.8839	0.0	0.57	0.904	0.2819	0.8839
9.4336	6.0	600	7.9411	0.3048	0.4533	0.3209	0.0	0.2202	0.3159	0.4649	0.7934	0.9024	0.0	0.76	0.914	0.3048	0.9024
8.996	7.0	700	7.2079	0.4942	0.6964	0.5359	0.0	0.2982	0.5086	0.5531	0.7995	0.9171	0.0	0.74	0.9305	0.4942	0.9171
8.3482	8.0	800	6.5042	0.4987	0.6906	0.5471	0.0	0.2387	0.5154	0.5768	0.8095	0.9095	0.0	0.75	0.922	0.4987	0.9095
7.9702	9.0	900	6.4708	0.5375	0.7037	0.6008	0.0	0.2762	0.5563	0.6024	0.8218	0.9209	0.0	0.78	0.9325	0.5375	0.9209
7.7992	10.0	1000	6.3745	0.5317	0.7162	0.5988	0.0	0.2473	0.551	0.5953	0.8299	0.9223	0.0	0.76	0.935	0.5317	0.9223

Framework versions

• Transformers 4.53.0.dev0
• Pytorch 2.7.1+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1