learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 300
num_epochs: 30

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 Artemia	Mar 100 Artemia
No log	1.0	250	10.0641	0.436	0.7974	0.4365	0.3445	0.5963	-1.0	0.3626	0.5701	0.6196	0.5457	0.7219	-1.0	0.436	0.6196
144.2508	2.0	500	8.1808	0.4542	0.8061	0.4576	0.3724	0.5804	-1.0	0.3642	0.5682	0.628	0.5715	0.708	-1.0	0.4542	0.628
144.2508	3.0	750	8.2538	0.5067	0.888	0.53	0.4411	0.594	-1.0	0.3869	0.5879	0.6349	0.5984	0.6861	-1.0	0.5067	0.6349
13.5443	4.0	1000	8.2240	0.489	0.8745	0.4725	0.4231	0.5799	-1.0	0.376	0.5769	0.6246	0.5753	0.6934	-1.0	0.489	0.6246
13.5443	5.0	1250	8.3286	0.4872	0.8734	0.5191	0.4191	0.5926	-1.0	0.3779	0.5769	0.6209	0.5731	0.6854	-1.0	0.4872	0.6209
11.9231	6.0	1500	8.2752	0.4927	0.8829	0.515	0.4207	0.5975	-1.0	0.3798	0.5866	0.6162	0.572	0.6774	-1.0	0.4927	0.6162
11.9231	7.0	1750	8.1550	0.4778	0.8845	0.4694	0.4079	0.5865	-1.0	0.3801	0.5869	0.6065	0.5484	0.6869	-1.0	0.4778	0.6065
10.8623	8.0	2000	8.1868	0.4851	0.8873	0.5028	0.417	0.5903	-1.0	0.3776	0.5826	0.6053	0.557	0.6723	-1.0	0.4851	0.6053
10.8623	9.0	2250	8.4927	0.4858	0.8852	0.4825	0.412	0.5917	-1.0	0.3819	0.5822	0.586	0.528	0.6657	-1.0	0.4858	0.586
9.8709	10.0	2500	8.5930	0.4643	0.8619	0.4697	0.3854	0.5831	-1.0	0.3682	0.5455	0.5483	0.4753	0.6496	-1.0	0.4643	0.5483
9.8709	11.0	2750	8.7453	0.4657	0.876	0.4324	0.3908	0.5819	-1.0	0.3769	0.5555	0.5595	0.4882	0.6584	-1.0	0.4657	0.5595
9.1291	12.0	3000	8.7313	0.4645	0.8695	0.4474	0.3905	0.5742	-1.0	0.3682	0.5414	0.5433	0.4667	0.6482	-1.0	0.4645	0.5433
9.1291	13.0	3250	9.2637	0.4583	0.8698	0.4557	0.3814	0.5787	-1.0	0.3648	0.5452	0.5458	0.4624	0.6606	-1.0	0.4583	0.5458
8.4350	14.0	3500	9.1713	0.4498	0.8413	0.441	0.3805	0.5725	-1.0	0.367	0.5492	0.5498	0.4801	0.6453	-1.0	0.4498	0.5498
8.4350	15.0	3750	9.2962	0.4601	0.863	0.4479	0.3886	0.5765	-1.0	0.3707	0.548	0.5486	0.4677	0.6606	-1.0	0.4601	0.5486
7.8666	16.0	4000	9.0432	0.4445	0.847	0.4204	0.3678	0.5681	-1.0	0.3604	0.5333	0.5333	0.4462	0.6526	-1.0	0.4445	0.5333
7.8666	17.0	4250	9.4077	0.4351	0.8298	0.4216	0.3538	0.5704	-1.0	0.3617	0.5349	0.5349	0.4532	0.6467	-1.0	0.4351	0.5349
7.2686	18.0	4500	9.4230	0.4471	0.8407	0.4349	0.3691	0.5768	-1.0	0.3626	0.5442	0.5442	0.4629	0.6555	-1.0	0.4471	0.5442
7.2686	19.0	4750	9.5215	0.4427	0.8436	0.4148	0.365	0.5791	-1.0	0.3558	0.5455	0.5455	0.4661	0.6547	-1.0	0.4427	0.5455
6.8319	20.0	5000	9.6549	0.4418	0.852	0.4199	0.3654	0.576	-1.0	0.3642	0.5411	0.5411	0.4597	0.6526	-1.0	0.4418	0.5411
6.8319	21.0	5250	10.0781	0.4357	0.8425	0.4311	0.3596	0.5684	-1.0	0.3548	0.5411	0.5411	0.4672	0.6431	-1.0	0.4357	0.5411
6.4622	22.0	5500	9.7193	0.4436	0.8308	0.4267	0.3649	0.581	-1.0	0.3539	0.5439	0.5439	0.4613	0.6577	-1.0	0.4436	0.5439
6.4622	23.0	5750	9.8768	0.4406	0.8281	0.4371	0.3634	0.577	-1.0	0.3573	0.5433	0.5433	0.4667	0.6489	-1.0	0.4406	0.5433
5.9281	24.0	6000	10.1613	0.4311	0.812	0.4351	0.3442	0.5828	-1.0	0.3576	0.5402	0.5402	0.4575	0.654	-1.0	0.4311	0.5402
5.9281	25.0	6250	10.2044	0.4267	0.8113	0.402	0.3435	0.5737	-1.0	0.3442	0.5364	0.5364	0.4575	0.6453	-1.0	0.4267	0.5364
5.6171	26.0	6500	10.5610	0.4297	0.8121	0.4276	0.3463	0.5838	-1.0	0.3477	0.5467	0.5467	0.4624	0.6628	-1.0	0.4297	0.5467
5.6171	27.0	6750	10.5552	0.4247	0.8036	0.4276	0.3415	0.5807	-1.0	0.3483	0.5405	0.5405	0.4591	0.6526	-1.0	0.4247	0.5405
5.2698	28.0	7000	10.5973	0.4322	0.8188	0.4228	0.3474	0.5831	-1.0	0.3542	0.5445	0.5445	0.464	0.6555	-1.0	0.4322	0.5445
5.2698	29.0	7250	10.6884	0.4284	0.8139	0.4229	0.3441	0.58	-1.0	0.3489	0.5421	0.5421	0.4624	0.6518	-1.0	0.4284	0.5421
4.9451	30.0	7500	10.7694	0.4268	0.8157	0.4215	0.3419	0.5803	-1.0	0.3455	0.5411	0.5411	0.4613	0.6511	-1.0	0.4268	0.5411

Framework versions

Transformers 5.9.0
Pytorch 2.8.0+cu128
Datasets 4.2.0
Tokenizers 0.22.2

dariacuna/rtdetr-v2-r101-finetune-28

rtdetr-v2-r101-finetune-28

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

dariacuna/
rtdetr-v2-r101-finetune-28