rtdetr-v2-r50-cppe5-finetune-2

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: 9.6769
• Map: 0.5368
• Map 50: 0.8312
• Map 75: 0.5962
• Map Small: 0.5364
• Map Medium: 0.441
• Map Large: 0.7689
• Mar 1: 0.3954
• Mar 10: 0.6567
• Mar 100: 0.6967
• Mar Small: 0.6067
• Mar Medium: 0.6153
• Mar Large: 0.8557
• Map Coverall: 0.5756
• Mar 100 Coverall: 0.7821
• Map Face Shield: 0.6521
• Mar 100 Face Shield: 0.8059
• Map Gloves: 0.4261
• Mar 100 Gloves: 0.5627
• Map Goggles: 0.4722
• Mar 100 Goggles: 0.6897
• Map Mask: 0.5578
• Mar 100 Mask: 0.6431

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 OptimizerNames.ADAMW_TORCH 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 Coverall	Mar 100 Coverall	Map Face Shield	Mar 100 Face Shield	Map Gloves	Mar 100 Gloves	Map Goggles	Mar 100 Goggles	Map Mask	Mar 100 Mask
No log	1.0	107	25.4682	0.0465	0.0795	0.0437	0.0028	0.0102	0.0671	0.0688	0.1912	0.2765	0.1031	0.1968	0.5097	0.2085	0.5748	0.002	0.243	0.0029	0.1746	0.001	0.1492	0.0184	0.2409
No log	2.0	214	15.4462	0.1823	0.3465	0.1617	0.0681	0.1115	0.2591	0.2151	0.4239	0.4884	0.2985	0.4232	0.7013	0.4591	0.6716	0.0879	0.5013	0.0658	0.3906	0.0495	0.4308	0.2489	0.4476
No log	3.0	321	12.7644	0.2555	0.4657	0.2476	0.0847	0.191	0.4466	0.2627	0.4528	0.5148	0.2473	0.4805	0.7435	0.5487	0.7185	0.1397	0.5468	0.1609	0.4183	0.1243	0.4138	0.3041	0.4764
No log	4.0	428	12.1356	0.2919	0.5558	0.2588	0.1466	0.2271	0.5099	0.285	0.4706	0.5293	0.3241	0.4886	0.729	0.5238	0.6986	0.2541	0.6101	0.1794	0.4281	0.1888	0.4508	0.3136	0.4587
36.0648	5.0	535	11.8591	0.3218	0.5856	0.2997	0.1436	0.2566	0.5379	0.3028	0.4821	0.5359	0.3055	0.5098	0.7267	0.5498	0.7104	0.2962	0.6139	0.1811	0.4147	0.2364	0.46	0.3456	0.4804
36.0648	6.0	642	11.7190	0.3157	0.5685	0.2997	0.1426	0.2557	0.5212	0.2823	0.4795	0.5398	0.3338	0.4924	0.7223	0.5437	0.7099	0.2111	0.5886	0.2288	0.4482	0.255	0.4677	0.34	0.4844
36.0648	7.0	749	11.9062	0.3212	0.5979	0.2974	0.1367	0.259	0.5387	0.2942	0.4765	0.5428	0.3495	0.5085	0.7153	0.5386	0.6905	0.303	0.6139	0.2067	0.4522	0.209	0.4754	0.3487	0.4818
36.0648	8.0	856	11.6933	0.3183	0.5969	0.2901	0.1414	0.2541	0.54	0.2977	0.483	0.5404	0.3387	0.4954	0.7399	0.5594	0.705	0.2812	0.6	0.2174	0.4429	0.208	0.4785	0.3253	0.4756
36.0648	9.0	963	11.6233	0.3202	0.5826	0.3226	0.1359	0.2583	0.5543	0.3035	0.4884	0.5462	0.3383	0.5126	0.7321	0.5472	0.6937	0.2746	0.619	0.2086	0.4545	0.2237	0.4754	0.347	0.4884
15.3215	10.0	1070	11.4090	0.3421	0.6207	0.3279	0.1389	0.2796	0.5757	0.316	0.4935	0.5477	0.3142	0.5067	0.745	0.5714	0.6991	0.3224	0.6278	0.2331	0.4371	0.2325	0.4754	0.3511	0.4991
15.3215	11.0	1177	11.5408	0.3436	0.6394	0.3212	0.1544	0.2759	0.5751	0.3175	0.5003	0.5554	0.3501	0.5119	0.7464	0.5444	0.7059	0.329	0.5949	0.2259	0.4571	0.2647	0.5108	0.354	0.5084
15.3215	12.0	1284	11.7707	0.3296	0.6169	0.3078	0.1525	0.2669	0.5336	0.3037	0.4793	0.5399	0.3164	0.4945	0.7323	0.5477	0.6892	0.32	0.6101	0.2295	0.429	0.21	0.4862	0.3408	0.4849
15.3215	13.0	1391	11.6683	0.3415	0.6231	0.3243	0.1447	0.2795	0.5746	0.3181	0.4946	0.5536	0.3625	0.5028	0.7387	0.5571	0.6901	0.3245	0.6114	0.2424	0.4585	0.2356	0.5092	0.3479	0.4987
15.3215	14.0	1498	11.7344	0.3305	0.6156	0.3133	0.1478	0.2877	0.5388	0.3194	0.4876	0.5468	0.3387	0.5232	0.7116	0.5236	0.7113	0.327	0.6	0.2255	0.4563	0.2373	0.4738	0.3391	0.4924
13.4858	15.0	1605	11.6264	0.3307	0.6056	0.3161	0.1213	0.2799	0.5711	0.3242	0.4933	0.5506	0.3454	0.5066	0.7434	0.5581	0.7144	0.3034	0.5962	0.2163	0.4701	0.2598	0.4862	0.3159	0.4862
13.4858	16.0	1712	11.5521	0.3287	0.6044	0.3125	0.1686	0.2751	0.5519	0.3171	0.4922	0.5484	0.3757	0.4978	0.7246	0.5635	0.7162	0.3018	0.5861	0.234	0.4714	0.236	0.48	0.3084	0.4884
13.4858	17.0	1819	11.7578	0.3382	0.6292	0.3237	0.164	0.281	0.5516	0.3215	0.4924	0.548	0.3353	0.5037	0.7302	0.5505	0.709	0.3225	0.6076	0.2187	0.4402	0.2704	0.5031	0.3286	0.48
13.4858	18.0	1926	11.5963	0.3454	0.6381	0.3218	0.1607	0.2921	0.5647	0.3294	0.4951	0.5507	0.3516	0.489	0.7565	0.5498	0.705	0.348	0.6051	0.2177	0.45	0.2613	0.5138	0.3504	0.4796
12.4151	19.0	2033	11.5293	0.3469	0.6347	0.3237	0.1415	0.2967	0.5694	0.323	0.4923	0.5415	0.3126	0.4894	0.733	0.5663	0.7032	0.345	0.5975	0.2389	0.4469	0.2667	0.4846	0.3175	0.4756
12.4151	20.0	2140	11.5551	0.3414	0.6306	0.3143	0.1716	0.2916	0.5768	0.3312	0.4969	0.5521	0.3257	0.513	0.7292	0.5629	0.7243	0.3179	0.5633	0.2498	0.4696	0.267	0.52	0.3095	0.4831
12.4151	21.0	2247	11.9833	0.3286	0.6184	0.2991	0.1597	0.277	0.533	0.3224	0.4898	0.5452	0.3502	0.5003	0.7228	0.5478	0.6955	0.2979	0.5899	0.2414	0.4638	0.2361	0.4923	0.3197	0.4844
12.4151	22.0	2354	11.9215	0.3408	0.6259	0.3184	0.142	0.2864	0.5548	0.3264	0.4893	0.5399	0.3216	0.4872	0.744	0.5429	0.6923	0.3578	0.619	0.2483	0.4585	0.2269	0.4569	0.3282	0.4729
12.4151	23.0	2461	12.0853	0.3304	0.6162	0.3031	0.1564	0.2852	0.5542	0.3198	0.4856	0.5275	0.309	0.4927	0.7118	0.5404	0.7041	0.3271	0.5886	0.242	0.4237	0.2325	0.4492	0.3097	0.472
11.6364	24.0	2568	11.8409	0.3344	0.622	0.3186	0.1689	0.2871	0.5485	0.3217	0.4938	0.5446	0.3457	0.4936	0.7208	0.5455	0.7126	0.2952	0.5899	0.2615	0.4638	0.2534	0.4862	0.3164	0.4707
11.6364	25.0	2675	12.1816	0.3201	0.5981	0.3021	0.1342	0.2717	0.5455	0.3151	0.4803	0.5303	0.3205	0.4817	0.7252	0.5315	0.7023	0.2957	0.5911	0.2163	0.4415	0.2379	0.4462	0.3188	0.4707
11.6364	26.0	2782	11.9448	0.3291	0.6113	0.2964	0.1687	0.2751	0.5635	0.3163	0.4875	0.5385	0.3434	0.4928	0.7221	0.5433	0.6919	0.2817	0.5797	0.2582	0.4746	0.2367	0.4708	0.3254	0.4756
11.6364	27.0	2889	11.9042	0.322	0.6094	0.2899	0.1286	0.2739	0.5564	0.3211	0.4919	0.5404	0.3371	0.4771	0.7404	0.5306	0.7005	0.3051	0.5975	0.2411	0.4509	0.228	0.4769	0.3052	0.4764
11.6364	28.0	2996	12.1391	0.3242	0.6003	0.3057	0.1342	0.2714	0.5507	0.3144	0.483	0.5308	0.3222	0.4825	0.7136	0.5356	0.6986	0.298	0.5848	0.2454	0.4487	0.2544	0.4538	0.2875	0.468
11.0017	29.0	3103	12.0627	0.3371	0.6166	0.3215	0.1445	0.2846	0.5479	0.3212	0.4874	0.5441	0.3267	0.5051	0.7284	0.5411	0.7077	0.3292	0.6038	0.2528	0.4451	0.2572	0.4877	0.3052	0.4764
11.0017	30.0	3210	12.3028	0.3353	0.6079	0.3192	0.158	0.2837	0.5625	0.315	0.4875	0.5332	0.2965	0.4917	0.7294	0.5347	0.6905	0.3534	0.6	0.246	0.4464	0.242	0.4662	0.3001	0.4631
11.0017	31.0	3317	11.9750	0.339	0.6148	0.325	0.1401	0.2827	0.5603	0.3195	0.4821	0.5328	0.2975	0.4866	0.7262	0.5451	0.7063	0.3469	0.5848	0.2502	0.4522	0.2412	0.4585	0.3115	0.4622
11.0017	32.0	3424	12.0644	0.3361	0.6158	0.3151	0.1374	0.2836	0.5539	0.3197	0.4886	0.5368	0.2821	0.5061	0.7212	0.5472	0.6982	0.3281	0.5886	0.2436	0.4612	0.2432	0.4615	0.3186	0.4747
10.4746	33.0	3531	11.9360	0.3323	0.615	0.3027	0.1597	0.2821	0.5495	0.3162	0.4863	0.5366	0.294	0.4971	0.7228	0.531	0.7032	0.3396	0.5949	0.2532	0.4705	0.2255	0.4477	0.3121	0.4667
10.4746	34.0	3638	11.7375	0.3393	0.6215	0.3145	0.1483	0.2853	0.5579	0.326	0.4915	0.5427	0.3199	0.5005	0.7224	0.5444	0.6968	0.3343	0.6076	0.2515	0.4638	0.2479	0.4615	0.3183	0.4836
10.4746	35.0	3745	11.9828	0.3282	0.605	0.3017	0.1392	0.2717	0.5518	0.3145	0.4801	0.5305	0.2918	0.4795	0.7182	0.5313	0.6973	0.3315	0.5835	0.2456	0.4616	0.2217	0.4446	0.3108	0.4653
10.4746	36.0	3852	11.8752	0.3302	0.6169	0.31	0.1597	0.2683	0.5583	0.3127	0.4814	0.5278	0.297	0.4793	0.709	0.5367	0.6968	0.3315	0.5848	0.2429	0.4402	0.2198	0.4477	0.32	0.4693
10.4746	37.0	3959	11.9312	0.3304	0.6097	0.3073	0.1444	0.2765	0.5464	0.3185	0.4809	0.536	0.3159	0.4909	0.7086	0.5382	0.6923	0.3265	0.6013	0.2545	0.454	0.213	0.4538	0.3198	0.4787
9.9527	38.0	4066	11.9053	0.3355	0.6135	0.3116	0.1443	0.283	0.5541	0.3188	0.4856	0.5333	0.3009	0.4804	0.7224	0.5382	0.6919	0.3493	0.5962	0.2481	0.4527	0.2271	0.4523	0.3151	0.4733
9.9527	39.0	4173	11.9321	0.331	0.6118	0.3094	0.1417	0.2754	0.5537	0.313	0.4817	0.5325	0.3038	0.4847	0.7139	0.538	0.6995	0.3312	0.5949	0.2491	0.4549	0.2289	0.4415	0.3079	0.4716
9.9527	40.0	4280	11.8940	0.3317	0.6135	0.3061	0.1432	0.276	0.5536	0.3136	0.4796	0.5283	0.2983	0.477	0.715	0.5378	0.7009	0.3298	0.5823	0.2506	0.4496	0.2308	0.4446	0.3095	0.464

Framework versions

• Transformers 4.52.3
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1