learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Use 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: 40

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	9.5500	0.3369	0.6354	0.3105	0.282	0.5106	-1.0	0.3424	0.6025	0.6779	0.629	0.7453	-1.0	0.3369	0.6779
17.0104	2.0	500	8.2829	0.4793	0.8595	0.4805	0.3963	0.5986	-1.0	0.3735	0.5888	0.6424	0.5887	0.7175	-1.0	0.4793	0.6424
17.0104	3.0	750	8.3439	0.4904	0.8608	0.4864	0.4082	0.604	-1.0	0.3816	0.605	0.6449	0.5823	0.7314	-1.0	0.4904	0.6449
13.7444	4.0	1000	8.0880	0.4859	0.8466	0.5024	0.3995	0.6051	-1.0	0.3779	0.5925	0.6159	0.5565	0.6971	-1.0	0.4859	0.6159
13.7444	5.0	1250	8.3034	0.4633	0.8579	0.4279	0.3732	0.5932	-1.0	0.3741	0.5844	0.629	0.5672	0.7139	-1.0	0.4633	0.629
12.4523	6.0	1500	8.3038	0.4459	0.8186	0.4326	0.3527	0.5881	-1.0	0.367	0.5869	0.6134	0.543	0.7088	-1.0	0.4459	0.6134
12.4523	7.0	1750	8.2535	0.4603	0.8475	0.4268	0.3745	0.5868	-1.0	0.3723	0.5807	0.5941	0.5172	0.7	-1.0	0.4603	0.5941
11.6434	8.0	2000	8.5828	0.4397	0.8158	0.4076	0.3479	0.5993	-1.0	0.3607	0.59	0.6199	0.5538	0.7109	-1.0	0.4397	0.6199
11.6434	9.0	2250	8.5738	0.4661	0.8492	0.4574	0.3831	0.5909	-1.0	0.3679	0.5891	0.6159	0.5462	0.7109	-1.0	0.4661	0.6159
10.838	10.0	2500	8.5913	0.4448	0.8092	0.3941	0.3512	0.6	-1.0	0.3583	0.5928	0.6112	0.5366	0.7139	-1.0	0.4448	0.6112
10.838	11.0	2750	8.3548	0.4493	0.8328	0.4192	0.3644	0.5865	-1.0	0.3626	0.5866	0.6034	0.5366	0.6971	-1.0	0.4493	0.6034
10.4572	12.0	3000	8.9282	0.3983	0.7714	0.3546	0.2969	0.5794	-1.0	0.3561	0.5754	0.5988	0.5183	0.708	-1.0	0.3983	0.5988
10.4572	13.0	3250	8.9473	0.4321	0.8265	0.3954	0.3458	0.5762	-1.0	0.3607	0.5745	0.5816	0.5194	0.6672	-1.0	0.4321	0.5816
9.8088	14.0	3500	9.5795	0.3875	0.7411	0.3588	0.2933	0.574	-1.0	0.3268	0.5769	0.5928	0.5258	0.6854	-1.0	0.3875	0.5928
9.8088	15.0	3750	8.7367	0.4284	0.8051	0.4216	0.3446	0.5768	-1.0	0.3489	0.5801	0.5857	0.5226	0.6737	-1.0	0.4284	0.5857
9.3282	16.0	4000	9.1113	0.4425	0.8309	0.3973	0.3621	0.5815	-1.0	0.3555	0.5751	0.5757	0.5075	0.6693	-1.0	0.4425	0.5757
9.3282	17.0	4250	9.0151	0.4281	0.7877	0.3922	0.332	0.5937	-1.0	0.3514	0.5807	0.5841	0.5086	0.6876	-1.0	0.4281	0.5841
8.6546	18.0	4500	9.0492	0.4203	0.8021	0.3741	0.3312	0.5717	-1.0	0.334	0.5579	0.5586	0.4833	0.662	-1.0	0.4203	0.5586
8.6546	19.0	4750	9.0114	0.4206	0.8081	0.3636	0.3281	0.5818	-1.0	0.3474	0.5592	0.5626	0.4812	0.6745	-1.0	0.4206	0.5626
8.1208	20.0	5000	9.4001	0.3837	0.7214	0.3456	0.2787	0.5878	-1.0	0.3349	0.5623	0.5626	0.4839	0.6708	-1.0	0.3837	0.5626
8.1208	21.0	5250	9.6652	0.4312	0.7938	0.3918	0.3341	0.5899	-1.0	0.3514	0.5601	0.5604	0.4742	0.6788	-1.0	0.4312	0.5604
7.6658	22.0	5500	9.4248	0.3954	0.7405	0.3525	0.2935	0.585	-1.0	0.3212	0.5492	0.5495	0.4629	0.6686	-1.0	0.3954	0.5495
7.6658	23.0	5750	9.6967	0.4077	0.7655	0.3818	0.311	0.5834	-1.0	0.3371	0.5704	0.5713	0.4978	0.6723	-1.0	0.4077	0.5713
7.1536	24.0	6000	9.6246	0.3909	0.7413	0.3623	0.293	0.582	-1.0	0.3287	0.5498	0.553	0.4763	0.6584	-1.0	0.3909	0.553
7.1536	25.0	6250	9.9725	0.3975	0.7365	0.3508	0.2934	0.5875	-1.0	0.3361	0.5486	0.5489	0.4624	0.6679	-1.0	0.3975	0.5489
6.8658	26.0	6500	10.5976	0.3705	0.6971	0.3271	0.268	0.5802	-1.0	0.3056	0.5439	0.5442	0.4581	0.6628	-1.0	0.3705	0.5442
6.8658	27.0	6750	9.9314	0.3819	0.7205	0.3516	0.282	0.5719	-1.0	0.3168	0.5399	0.5399	0.4591	0.6511	-1.0	0.3819	0.5399
6.5861	28.0	7000	10.8218	0.3784	0.7138	0.3428	0.2783	0.5811	-1.0	0.3283	0.5573	0.5576	0.4828	0.6606	-1.0	0.3784	0.5576
6.5861	29.0	7250	10.3239	0.373	0.6991	0.3375	0.2699	0.5803	-1.0	0.3121	0.5576	0.5579	0.4833	0.6599	-1.0	0.373	0.5579
6.1734	30.0	7500	10.1381	0.3875	0.725	0.354	0.2803	0.5903	-1.0	0.3181	0.5489	0.5492	0.4656	0.6635	-1.0	0.3875	0.5492
6.1734	31.0	7750	10.5858	0.3696	0.6893	0.3462	0.2621	0.5849	-1.0	0.3118	0.5483	0.5486	0.4645	0.6635	-1.0	0.3696	0.5486
5.825	32.0	8000	10.3701	0.3893	0.7292	0.3454	0.2854	0.5887	-1.0	0.3308	0.5495	0.5495	0.4694	0.6599	-1.0	0.3893	0.5495
5.825	33.0	8250	10.7725	0.3731	0.6927	0.3423	0.2688	0.5811	-1.0	0.3305	0.5533	0.5533	0.4758	0.6599	-1.0	0.3731	0.5533
5.571	34.0	8500	10.6223	0.3745	0.6969	0.3358	0.2683	0.5824	-1.0	0.315	0.5442	0.5442	0.457	0.6642	-1.0	0.3745	0.5442
5.571	35.0	8750	10.7401	0.3783	0.7084	0.3419	0.2723	0.5883	-1.0	0.3181	0.5489	0.5489	0.4677	0.6606	-1.0	0.3783	0.5489
5.2802	36.0	9000	11.1337	0.3769	0.699	0.3483	0.2715	0.586	-1.0	0.3234	0.5517	0.5517	0.4704	0.6642	-1.0	0.3769	0.5517
5.2802	37.0	9250	11.1612	0.3734	0.6936	0.3425	0.2651	0.5843	-1.0	0.3187	0.5498	0.5498	0.4683	0.662	-1.0	0.3734	0.5498
5.0324	38.0	9500	11.3974	0.3653	0.6799	0.3413	0.2528	0.5877	-1.0	0.3128	0.547	0.547	0.4629	0.662	-1.0	0.3653	0.547
5.0324	39.0	9750	11.5781	0.3672	0.6814	0.3499	0.2555	0.5897	-1.0	0.3146	0.5489	0.5489	0.464	0.6657	-1.0	0.3672	0.5489
4.7734	40.0	10000	11.5671	0.3661	0.6783	0.3478	0.2545	0.5879	-1.0	0.3134	0.5483	0.5483	0.4645	0.6635	-1.0	0.3661	0.5483

Framework versions

Transformers 4.57.1
Pytorch 2.8.0+cu128
Datasets 4.2.0
Tokenizers 0.22.1

dariacuna/rtdetr-v2-r50-finetune-8

rtdetr-v2-r50-finetune-8

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

dariacuna/
rtdetr-v2-r50-finetune-8