rtdetr-v2-r50-finetune-6

This model is a fine-tuned version of PekingU/rtdetr_v2_r50vd on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 7.8917
Map: 0.5563
Map 50: 0.8997
Map 75: 0.6605
Map Small: 0.5182
Map Medium: 0.6443
Map Large: -1.0
Mar 1: 0.3027
Mar 10: 0.627
Mar 100: 0.6805
Mar Small: 0.64
Mar Medium: 0.7554
Mar Large: -1.0

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: 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: 60

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
No log	1.0	69	79.2112	0.0008	0.0039	0.0	0.0008	0.0017	-1.0	0.0	0.0158	0.1177	0.0904	0.1637	-1.0
No log	2.0	138	25.9251	0.2511	0.4688	0.225	0.1597	0.4305	-1.0	0.26	0.5377	0.5633	0.4607	0.7362	-1.0
No log	3.0	207	14.0285	0.438	0.8012	0.4587	0.3637	0.5608	-1.0	0.3484	0.5391	0.594	0.5237	0.7125	-1.0
No log	4.0	276	10.6173	0.48	0.8319	0.5228	0.413	0.6022	-1.0	0.3772	0.5595	0.6009	0.5326	0.7163	-1.0
No log	5.0	345	9.6359	0.4888	0.8318	0.5267	0.4207	0.6104	-1.0	0.3763	0.5577	0.6233	0.5526	0.7425	-1.0
No log	6.0	414	9.2794	0.483	0.8243	0.5373	0.4177	0.6033	-1.0	0.3735	0.5535	0.6135	0.5393	0.7387	-1.0
No log	7.0	483	9.1721	0.4893	0.8441	0.566	0.424	0.611	-1.0	0.3842	0.5605	0.6112	0.5407	0.73	-1.0
54.7379	8.0	552	9.2472	0.4701	0.8219	0.4962	0.4079	0.5932	-1.0	0.3749	0.547	0.6233	0.5533	0.7412	-1.0
54.7379	9.0	621	9.2285	0.4724	0.8101	0.522	0.4129	0.5867	-1.0	0.3819	0.5419	0.5986	0.523	0.7262	-1.0
54.7379	10.0	690	9.2690	0.4555	0.8175	0.4937	0.39	0.5788	-1.0	0.373	0.5298	0.5981	0.5281	0.7163	-1.0
54.7379	11.0	759	9.2237	0.4694	0.8232	0.4771	0.411	0.5857	-1.0	0.3712	0.5437	0.5865	0.5237	0.6925	-1.0
54.7379	12.0	828	9.0750	0.47	0.8093	0.5023	0.4136	0.581	-1.0	0.3809	0.5367	0.5702	0.5015	0.6862	-1.0
54.7379	13.0	897	9.1644	0.4661	0.8146	0.4899	0.4113	0.5769	-1.0	0.3749	0.5391	0.5758	0.5104	0.6862	-1.0
54.7379	14.0	966	9.2909	0.4632	0.8001	0.4825	0.4061	0.5759	-1.0	0.3688	0.5312	0.5605	0.4896	0.68	-1.0
10.9263	15.0	1035	9.4030	0.4713	0.8156	0.5057	0.4166	0.5755	-1.0	0.3814	0.5419	0.5795	0.5126	0.6925	-1.0
10.9263	16.0	1104	9.3454	0.4679	0.8195	0.5017	0.4163	0.5618	-1.0	0.3749	0.5395	0.5605	0.5022	0.6587	-1.0
10.9263	17.0	1173	9.4194	0.47	0.817	0.5033	0.4156	0.5771	-1.0	0.3707	0.5391	0.5595	0.4881	0.68	-1.0
10.9263	18.0	1242	9.9727	0.4573	0.8201	0.4312	0.4003	0.5709	-1.0	0.3674	0.5428	0.5633	0.4956	0.6775	-1.0
10.9263	19.0	1311	9.5181	0.4702	0.8244	0.5254	0.4235	0.5634	-1.0	0.374	0.5428	0.5572	0.4889	0.6725	-1.0
10.9263	20.0	1380	10.0705	0.4686	0.8269	0.4865	0.4143	0.5742	-1.0	0.3758	0.5526	0.573	0.5022	0.6925	-1.0
10.9263	21.0	1449	9.4824	0.4811	0.8333	0.5139	0.4275	0.5803	-1.0	0.3847	0.5484	0.5558	0.4941	0.66	-1.0
8.529	22.0	1518	9.8264	0.4742	0.8323	0.4832	0.4248	0.5684	-1.0	0.3758	0.5442	0.5581	0.4881	0.6762	-1.0
8.529	23.0	1587	9.8779	0.4792	0.8171	0.5289	0.4212	0.5834	-1.0	0.3791	0.5437	0.547	0.4763	0.6662	-1.0
8.529	24.0	1656	9.9106	0.4776	0.8266	0.5265	0.4267	0.5736	-1.0	0.3805	0.5447	0.5474	0.4837	0.655	-1.0
8.529	25.0	1725	10.0986	0.4811	0.8339	0.4607	0.4269	0.5799	-1.0	0.3767	0.5437	0.5488	0.4756	0.6725	-1.0
8.529	26.0	1794	9.6901	0.4814	0.8327	0.5142	0.4242	0.5813	-1.0	0.3814	0.5419	0.5465	0.4778	0.6625	-1.0
8.529	27.0	1863	10.4926	0.483	0.8316	0.5255	0.4231	0.5887	-1.0	0.3777	0.5433	0.5493	0.4778	0.67	-1.0
8.529	28.0	1932	9.9321	0.4773	0.8227	0.5106	0.4263	0.5736	-1.0	0.3809	0.5419	0.5437	0.48	0.6513	-1.0
7.0744	29.0	2001	9.8464	0.4758	0.8366	0.5207	0.4221	0.5731	-1.0	0.3781	0.5409	0.5433	0.4785	0.6525	-1.0
7.0744	30.0	2070	9.8972	0.4846	0.8336	0.5556	0.4382	0.5703	-1.0	0.3809	0.5433	0.5451	0.4867	0.6438	-1.0
7.0744	31.0	2139	9.9210	0.4774	0.8264	0.4882	0.4279	0.5726	-1.0	0.3805	0.5405	0.5437	0.4793	0.6525	-1.0
7.0744	32.0	2208	10.2300	0.4746	0.818	0.513	0.4264	0.5647	-1.0	0.3777	0.5386	0.5391	0.477	0.6438	-1.0
7.0744	33.0	2277	9.9653	0.4795	0.8327	0.4958	0.4278	0.5717	-1.0	0.3791	0.5405	0.5419	0.4748	0.655	-1.0
7.0744	34.0	2346	10.5432	0.4827	0.8304	0.5129	0.4295	0.578	-1.0	0.387	0.5442	0.5447	0.4785	0.6562	-1.0
7.0744	35.0	2415	10.3663	0.4773	0.8252	0.5095	0.422	0.5754	-1.0	0.3744	0.5405	0.5414	0.4733	0.6562	-1.0
7.0744	36.0	2484	10.5355	0.4756	0.8213	0.4972	0.4238	0.572	-1.0	0.38	0.5395	0.5423	0.4741	0.6575	-1.0
6.1286	37.0	2553	10.2664	0.4768	0.8263	0.5364	0.4235	0.5724	-1.0	0.3791	0.5419	0.5437	0.477	0.6562	-1.0
6.1286	38.0	2622	10.3030	0.4834	0.8299	0.5608	0.4322	0.5729	-1.0	0.3823	0.5465	0.5479	0.4859	0.6525	-1.0
6.1286	39.0	2691	10.3299	0.4817	0.8347	0.5121	0.4286	0.5777	-1.0	0.3791	0.5442	0.547	0.4822	0.6562	-1.0
6.1286	40.0	2760	10.9503	0.4781	0.8257	0.4964	0.4284	0.5723	-1.0	0.3781	0.54	0.5433	0.4748	0.6587	-1.0
6.1286	41.0	2829	10.5873	0.4795	0.8261	0.5319	0.4325	0.5684	-1.0	0.3805	0.5423	0.5456	0.4837	0.65	-1.0
6.1286	42.0	2898	10.9400	0.4809	0.8288	0.537	0.435	0.5676	-1.0	0.3828	0.5428	0.5456	0.4822	0.6525	-1.0
6.1286	43.0	2967	10.5762	0.4799	0.8302	0.5494	0.4315	0.5677	-1.0	0.3814	0.5419	0.5447	0.4852	0.645	-1.0
5.383	44.0	3036	10.6760	0.481	0.8299	0.526	0.4299	0.5712	-1.0	0.3795	0.5405	0.5437	0.4822	0.6475	-1.0
5.383	45.0	3105	10.8686	0.4849	0.8333	0.5066	0.4374	0.5726	-1.0	0.386	0.5456	0.5479	0.483	0.6575	-1.0
5.383	46.0	3174	10.7009	0.4823	0.8344	0.5332	0.4326	0.5708	-1.0	0.3795	0.5442	0.547	0.4844	0.6525	-1.0
5.383	47.0	3243	10.9353	0.4845	0.8407	0.5178	0.4336	0.5768	-1.0	0.3842	0.5433	0.546	0.4807	0.6562	-1.0
5.383	48.0	3312	10.9563	0.4844	0.8307	0.5213	0.4372	0.5711	-1.0	0.386	0.5456	0.5493	0.4859	0.6562	-1.0
5.383	49.0	3381	10.9962	0.4806	0.8292	0.5367	0.431	0.5693	-1.0	0.3809	0.5419	0.5433	0.48	0.65	-1.0
5.383	50.0	3450	11.2143	0.4838	0.8346	0.5303	0.4364	0.5748	-1.0	0.3847	0.5456	0.5479	0.4837	0.6562	-1.0
4.726	51.0	3519	11.1295	0.4836	0.83	0.5252	0.4324	0.5778	-1.0	0.3814	0.5433	0.5447	0.477	0.6587	-1.0
4.726	52.0	3588	10.8141	0.4782	0.8331	0.5215	0.4283	0.5691	-1.0	0.3833	0.54	0.5428	0.48	0.6488	-1.0
4.726	53.0	3657	11.3813	0.4852	0.8304	0.5232	0.4335	0.5773	-1.0	0.3842	0.5442	0.547	0.4815	0.6575	-1.0
4.726	54.0	3726	11.4009	0.4812	0.8342	0.5314	0.4343	0.5676	-1.0	0.3823	0.5428	0.5447	0.4815	0.6513	-1.0
4.726	55.0	3795	11.4553	0.4864	0.8313	0.5368	0.4357	0.5767	-1.0	0.3847	0.5451	0.5474	0.4815	0.6587	-1.0
4.726	56.0	3864	11.5131	0.4829	0.8312	0.5186	0.4346	0.5711	-1.0	0.3842	0.5428	0.5451	0.4815	0.6525	-1.0
4.726	57.0	3933	11.5599	0.4811	0.8316	0.5209	0.4304	0.5722	-1.0	0.3814	0.5423	0.5447	0.4793	0.655	-1.0
4.2483	58.0	4002	11.4917	0.4838	0.8314	0.5214	0.4334	0.5744	-1.0	0.3851	0.5437	0.546	0.48	0.6575	-1.0
4.2483	59.0	4071	11.6793	0.4839	0.8307	0.5205	0.4326	0.5756	-1.0	0.3847	0.5451	0.546	0.4785	0.66	-1.0
4.2483	60.0	4140	11.6224	0.4845	0.8307	0.5263	0.4342	0.5735	-1.0	0.3851	0.5451	0.5479	0.483	0.6575	-1.0

Framework versions

Transformers 4.57.1
Pytorch 2.8.0+cu128
Datasets 4.2.0
Tokenizers 0.22.1

dariacuna/rtdetr-v2-r50-finetune-6

rtdetr-v2-r50-finetune-6

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Seed this model