rtdetr-paper-control-300ep-20260305-v1

This model is a fine-tuned version of PekingU/rtdetr_r101vd_coco_o365 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 7.3432
Map: 0.4797
Map 50: 0.7782
Map 75: 0.5303
Map Small: 0.0505
Map Medium: 0.4796
Map Large: 0.56
Mar 1: 0.0571
Mar 10: 0.3444
Mar 100: 0.6572
Mar Small: 0.05
Mar Medium: 0.6488
Mar Large: 0.8162
Map Tray: 0.4304
Mar 100 Tray: 0.5838
Map Cart: 0.529
Mar 100 Cart: 0.7306

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: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
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: constant
num_epochs: 300
mixed_precision_training: Native AMP

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 Tray	Mar 100 Tray	Map Cart	Mar 100 Cart
No log	1.0	7	52.0209	0.0002	0.0005	0.0	0.0	0.0004	0.0007	0.0	0.0	0.0256	0.0	0.0225	0.0843	0.0	0.0029	0.0003	0.0484
No log	2.0	14	31.5488	0.0103	0.0265	0.0072	0.0	0.012	0.0227	0.001	0.0176	0.0693	0.0	0.0629	0.1912	0.02	0.0886	0.0005	0.05
No log	3.0	21	28.2528	0.0127	0.0338	0.0072	0.0	0.0097	0.0421	0.0012	0.0177	0.1023	0.0	0.0906	0.3206	0.0242	0.0998	0.0012	0.1048
No log	4.0	28	37.8709	0.0162	0.0515	0.0056	0.0	0.0159	0.0494	0.0007	0.0313	0.0762	0.0	0.0696	0.1971	0.0309	0.104	0.0015	0.0484
No log	5.0	35	37.3820	0.0346	0.1046	0.0173	0.0	0.0353	0.0477	0.0021	0.0411	0.1068	0.0	0.1022	0.1946	0.0655	0.1475	0.0038	0.0661
No log	6.0	42	36.1954	0.0651	0.1492	0.0425	0.0168	0.0656	0.0948	0.0134	0.0581	0.1476	0.05	0.1439	0.223	0.1147	0.2242	0.0155	0.071
No log	7.0	49	30.6148	0.0861	0.1573	0.0883	0.0404	0.0914	0.1376	0.0103	0.078	0.1941	0.2	0.1876	0.3127	0.1555	0.2881	0.0168	0.1
No log	8.0	56	23.8829	0.1194	0.2319	0.1211	0.101	0.1289	0.1371	0.0193	0.1082	0.2528	0.1	0.2396	0.501	0.1878	0.3475	0.051	0.1581
No log	9.0	63	19.4596	0.1697	0.3124	0.1718	0.2427	0.1784	0.1977	0.0266	0.1436	0.3223	0.35	0.3084	0.5701	0.2327	0.3961	0.1067	0.2484
No log	10.0	70	15.2607	0.1917	0.3932	0.1882	0.2427	0.1888	0.3115	0.0362	0.1644	0.378	0.35	0.3633	0.6422	0.2132	0.4061	0.1702	0.35
No log	11.0	77	14.7306	0.2121	0.4	0.225	0.3515	0.2133	0.3162	0.0326	0.1666	0.3806	0.35	0.3747	0.4784	0.2501	0.4353	0.174	0.3258
No log	12.0	84	13.5177	0.1856	0.3542	0.1942	0.0	0.1853	0.2909	0.0224	0.1383	0.3419	0.0	0.3274	0.6181	0.2646	0.4531	0.1065	0.2306
No log	13.0	91	12.6535	0.2237	0.4062	0.2263	0.0	0.2307	0.3097	0.0438	0.1653	0.383	0.0	0.3669	0.6892	0.2829	0.4515	0.1645	0.3145
No log	14.0	98	12.0757	0.2536	0.438	0.2623	0.0	0.2531	0.3565	0.0506	0.1969	0.4104	0.0	0.3942	0.7221	0.3114	0.4966	0.1959	0.3242
No log	15.0	105	11.2874	0.233	0.4118	0.2445	0.0	0.2371	0.3383	0.0364	0.1578	0.3956	0.0	0.3819	0.6627	0.3666	0.5525	0.0994	0.2387
No log	16.0	112	10.8548	0.2832	0.4823	0.2996	0.0	0.2847	0.3631	0.0447	0.2115	0.4504	0.0	0.4371	0.7157	0.3939	0.5557	0.1726	0.3452
No log	17.0	119	10.5296	0.3039	0.5178	0.325	0.0	0.31	0.3528	0.0473	0.2097	0.448	0.0	0.4408	0.5887	0.3865	0.5541	0.2213	0.3419
No log	18.0	126	9.9201	0.3223	0.5444	0.3532	0.101	0.3304	0.3149	0.0541	0.227	0.4706	0.1	0.4641	0.5951	0.4061	0.5669	0.2386	0.3742
No log	19.0	133	10.3674	0.3342	0.5783	0.349	0.101	0.3423	0.3409	0.0556	0.237	0.4733	0.1	0.464	0.651	0.4026	0.5433	0.2658	0.4032
No log	20.0	140	9.8158	0.3266	0.5346	0.3411	0.101	0.3375	0.3533	0.0505	0.2243	0.4896	0.1	0.4798	0.6779	0.4056	0.5711	0.2475	0.4081
No log	21.0	147	9.5132	0.3562	0.5817	0.3936	0.101	0.3662	0.379	0.0521	0.2586	0.4988	0.1	0.4878	0.7127	0.4309	0.5831	0.2814	0.4145
No log	22.0	154	9.4140	0.368	0.6068	0.4151	0.101	0.3649	0.4702	0.0526	0.2695	0.5164	0.1	0.5008	0.8172	0.418	0.573	0.3179	0.4597
No log	23.0	161	9.2565	0.3727	0.623	0.3976	0.101	0.3734	0.4767	0.054	0.2727	0.5187	0.1	0.5045	0.7946	0.4244	0.5665	0.321	0.471
No log	24.0	168	8.9574	0.3983	0.6546	0.4503	0.101	0.4023	0.4551	0.0524	0.2917	0.5378	0.1	0.5212	0.8613	0.4353	0.5676	0.3613	0.5081
No log	25.0	175	8.5775	0.4209	0.6776	0.4786	0.101	0.4178	0.5274	0.06	0.303	0.5542	0.1	0.5403	0.8245	0.4316	0.5923	0.4101	0.5161
No log	26.0	182	8.3658	0.4661	0.7476	0.5103	0.101	0.465	0.5367	0.0638	0.3461	0.6212	0.1	0.6119	0.8005	0.4063	0.5843	0.526	0.6581
No log	27.0	189	8.1695	0.4829	0.7707	0.5387	0.101	0.4884	0.509	0.0584	0.3327	0.637	0.1	0.6282	0.8049	0.413	0.5852	0.5528	0.6887
No log	28.0	196	8.4744	0.468	0.7682	0.5038	0.101	0.4691	0.5257	0.0603	0.3253	0.6179	0.1	0.6072	0.8211	0.3953	0.5665	0.5407	0.6694
No log	29.0	203	7.9397	0.5061	0.8003	0.5454	0.101	0.5051	0.5746	0.0592	0.3478	0.655	0.1	0.6465	0.8181	0.4259	0.5907	0.5863	0.7194
No log	30.0	210	7.9048	0.5079	0.8185	0.5428	0.101	0.5063	0.5854	0.0619	0.3495	0.6593	0.1	0.651	0.8167	0.4138	0.5799	0.602	0.7387
No log	31.0	217	8.0720	0.4907	0.7843	0.5395	0.101	0.4981	0.4826	0.055	0.3456	0.6649	0.1	0.658	0.7956	0.4155	0.5878	0.5659	0.7419
No log	32.0	224	8.0961	0.5034	0.8035	0.5719	0.101	0.5012	0.6458	0.0621	0.3407	0.655	0.1	0.6455	0.8328	0.393	0.5713	0.6139	0.7387
No log	33.0	231	7.9472	0.5262	0.8299	0.6003	0.101	0.5239	0.671	0.0684	0.342	0.6618	0.1	0.6543	0.8044	0.4309	0.5961	0.6214	0.7274
No log	34.0	238	7.8616	0.5284	0.8342	0.5978	0.101	0.5267	0.6558	0.0659	0.3517	0.6733	0.1	0.6656	0.8196	0.4233	0.5886	0.6336	0.7581
No log	35.0	245	7.9081	0.5432	0.8471	0.6349	0.101	0.5417	0.671	0.0649	0.3569	0.6758	0.1	0.6688	0.8088	0.4337	0.5936	0.6526	0.7581
No log	36.0	252	8.1714	0.5404	0.8545	0.6202	0.101	0.5407	0.6592	0.0642	0.3552	0.6631	0.1	0.6549	0.8162	0.4375	0.5811	0.6434	0.7452
No log	37.0	259	7.8066	0.5435	0.841	0.6259	0.101	0.5436	0.5881	0.0617	0.3531	0.6645	0.1	0.6554	0.8392	0.4407	0.5904	0.6463	0.7387
No log	38.0	266	7.8896	0.5251	0.8331	0.6025	0.101	0.5266	0.5869	0.056	0.3511	0.6608	0.1	0.6522	0.8225	0.428	0.586	0.6223	0.7355
No log	39.0	273	7.6938	0.5327	0.8474	0.6161	0.101	0.5324	0.6111	0.0575	0.3509	0.6528	0.1	0.6419	0.8588	0.4409	0.5814	0.6246	0.7242
No log	40.0	280	7.7116	0.5442	0.8495	0.6457	0.101	0.5459	0.6057	0.0579	0.3562	0.6613	0.1	0.6528	0.824	0.4645	0.6	0.6238	0.7226
No log	41.0	287	7.8024	0.5437	0.8543	0.6388	0.101	0.5458	0.5457	0.062	0.3529	0.6596	0.1	0.6499	0.8422	0.4521	0.5852	0.6354	0.7339
No log	42.0	294	7.6427	0.5433	0.8435	0.6397	0.101	0.5472	0.5426	0.0521	0.3493	0.6672	0.1	0.6573	0.8559	0.4678	0.6053	0.6189	0.729
No log	43.0	301	7.5489	0.5314	0.8411	0.6176	0.101	0.5333	0.5614	0.0536	0.3518	0.6603	0.1	0.6506	0.8436	0.4485	0.5915	0.6144	0.729
No log	44.0	308	7.5076	0.5344	0.8317	0.6286	0.101	0.5371	0.5468	0.0601	0.3573	0.668	0.1	0.6589	0.8422	0.4452	0.5989	0.6237	0.7371
No log	45.0	315	7.7148	0.5219	0.8304	0.6008	0.101	0.5246	0.5515	0.0571	0.3494	0.6583	0.1	0.6486	0.8436	0.4332	0.586	0.6106	0.7306
No log	46.0	322	7.6162	0.5108	0.8174	0.5738	0.101	0.5105	0.5635	0.0594	0.3511	0.6538	0.1	0.644	0.8387	0.4182	0.5851	0.6034	0.7226
No log	47.0	329	7.5197	0.5304	0.8283	0.6041	0.101	0.529	0.6265	0.0599	0.3575	0.6725	0.1	0.6625	0.8632	0.4452	0.5966	0.6156	0.7484
No log	48.0	336	7.5130	0.5234	0.8292	0.5979	0.3515	0.5208	0.6144	0.0604	0.3587	0.6655	0.35	0.6548	0.8603	0.4321	0.5907	0.6148	0.7403
No log	49.0	343	7.9138	0.4962	0.7833	0.5739	0.101	0.5032	0.4042	0.0593	0.3344	0.6418	0.1	0.6306	0.851	0.4044	0.5626	0.5879	0.721
No log	50.0	350	7.5024	0.5342	0.8298	0.6249	0.101	0.5336	0.5962	0.0606	0.3548	0.6695	0.1	0.6599	0.851	0.4386	0.5955	0.6297	0.7435
No log	51.0	357	7.5145	0.4965	0.7934	0.5972	0.0505	0.4976	0.5537	0.0621	0.3492	0.6605	0.05	0.6502	0.8554	0.4376	0.5807	0.5553	0.7403
No log	52.0	364	7.4957	0.5143	0.8056	0.6085	0.101	0.5138	0.5452	0.0579	0.3551	0.6725	0.1	0.6637	0.8387	0.4315	0.5982	0.5971	0.7468
No log	53.0	371	7.5725	0.5109	0.8051	0.5893	0.101	0.511	0.5503	0.0596	0.3524	0.662	0.1	0.6519	0.8539	0.4193	0.5966	0.6025	0.7274
No log	54.0	378	7.5533	0.5233	0.8235	0.6034	0.101	0.5218	0.6023	0.0652	0.3531	0.6598	0.1	0.6507	0.8314	0.425	0.5921	0.6215	0.7274
No log	55.0	385	7.3714	0.5347	0.8371	0.613	0.101	0.5345	0.5914	0.061	0.3596	0.6711	0.1	0.6633	0.8176	0.4538	0.6019	0.6156	0.7403
No log	56.0	392	7.4263	0.5078	0.8067	0.584	0.0505	0.5084	0.5391	0.0592	0.3572	0.6642	0.05	0.6544	0.851	0.4286	0.5994	0.587	0.729
No log	57.0	399	7.2185	0.5168	0.8143	0.6013	0.0505	0.5177	0.5529	0.0629	0.356	0.6768	0.05	0.6687	0.8314	0.4255	0.6003	0.608	0.7532
No log	58.0	406	7.4327	0.5198	0.8265	0.6135	0.0505	0.5219	0.5466	0.0619	0.3564	0.6693	0.05	0.6606	0.8343	0.4229	0.5902	0.6167	0.7484
No log	59.0	413	7.2584	0.5303	0.8424	0.6396	0.101	0.5293	0.6028	0.0616	0.3479	0.6725	0.1	0.663	0.8525	0.4445	0.603	0.6161	0.7419
No log	60.0	420	7.3321	0.5164	0.8176	0.609	0.0505	0.5159	0.5792	0.0643	0.3565	0.6633	0.05	0.6534	0.8525	0.4451	0.5944	0.5878	0.7323
No log	61.0	427	7.4019	0.5209	0.8239	0.6145	0.0505	0.5205	0.6069	0.0588	0.3542	0.6601	0.05	0.6529	0.7966	0.4429	0.5928	0.599	0.7274
No log	62.0	434	7.2498	0.5257	0.8223	0.6158	0.0505	0.5253	0.5705	0.0617	0.3581	0.6717	0.05	0.664	0.8176	0.4421	0.595	0.6094	0.7484
No log	63.0	441	7.3263	0.5094	0.806	0.6249	0.0505	0.5089	0.5569	0.0638	0.3569	0.6582	0.05	0.6485	0.8436	0.4176	0.5905	0.6013	0.7258
No log	64.0	448	7.3728	0.5101	0.8177	0.5958	0.0505	0.5099	0.5888	0.0643	0.3524	0.6603	0.05	0.651	0.8358	0.4154	0.5819	0.6049	0.7387
No log	65.0	455	7.4039	0.5064	0.8135	0.5825	0.2515	0.5069	0.5459	0.0575	0.3511	0.6587	0.25	0.6503	0.8132	0.4237	0.5917	0.5891	0.7258
No log	66.0	462	7.3934	0.5249	0.8203	0.6113	0.3515	0.5252	0.5033	0.0643	0.3561	0.6694	0.35	0.6601	0.8392	0.4341	0.6034	0.6156	0.7355
No log	67.0	469	7.5553	0.4979	0.8092	0.5538	0.2515	0.4984	0.543	0.0639	0.3498	0.6533	0.25	0.6448	0.8074	0.4112	0.5743	0.5846	0.7323
No log	68.0	476	7.4750	0.523	0.8388	0.6028	0.101	0.5258	0.4703	0.0529	0.3469	0.6538	0.1	0.6444	0.8314	0.4536	0.5883	0.5923	0.7194
No log	69.0	483	7.3127	0.5208	0.8292	0.5892	0.101	0.5198	0.6035	0.06	0.3498	0.6571	0.1	0.6494	0.8029	0.4318	0.59	0.6098	0.7242
No log	70.0	490	7.3432	0.4797	0.7782	0.5303	0.0505	0.4796	0.56	0.0571	0.3444	0.6572	0.05	0.6488	0.8162	0.4304	0.5838	0.529	0.7306

Framework versions

Transformers 5.3.0
Pytorch 2.10.0+cu128
Datasets 4.6.1
Tokenizers 0.22.2

nielsr/rtdetr-paper-control-300ep-20260305-v1

rtdetr-paper-control-300ep-20260305-v1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Seed this model