PekingU_rtdetr_r50vd_cppe5_jw_1

This model is a fine-tuned version of PekingU/rtdetr_r50vd on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 12.2150
• Map: 0.3305
• Map 50: 0.6332
• Map 75: 0.2924
• Map Small: 0.1087
• Map Medium: 0.2627
• Map Large: 0.5123
• Mar 1: 0.3018
• Mar 10: 0.4833
• Mar 100: 0.5384
• Mar Small: 0.3203
• Mar Medium: 0.478
• Mar Large: 0.7067
• Map Coverall: 0.5473
• Mar 100 Coverall: 0.7158
• Map Face Shield: 0.3566
• Mar 100 Face Shield: 0.6063
• Map Gloves: 0.2415
• Mar 100 Gloves: 0.4621
• Map Goggles: 0.2167
• Mar 100 Goggles: 0.4431
• Map Mask: 0.2905
• Mar 100 Mask: 0.4649

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• 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 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	13.6921	0.2046	0.3735	0.1947	0.068	0.1841	0.3025	0.2454	0.4315	0.4921	0.3014	0.4276	0.6762	0.4274	0.6712	0.0666	0.4848	0.116	0.3835	0.0986	0.4277	0.3143	0.4933
No log	2.0	214	12.2598	0.2507	0.4883	0.2181	0.1039	0.2171	0.3973	0.2841	0.4745	0.537	0.3358	0.4862	0.6937	0.4926	0.6901	0.1416	0.5722	0.1704	0.4478	0.1357	0.4769	0.3134	0.4982
No log	3.0	321	11.9174	0.2813	0.5306	0.2637	0.0892	0.2413	0.4566	0.2967	0.5005	0.5551	0.3245	0.5093	0.7123	0.5103	0.7081	0.214	0.6076	0.1968	0.4679	0.1747	0.4769	0.3108	0.5151
No log	4.0	428	11.9953	0.3027	0.5694	0.2825	0.1123	0.2568	0.4667	0.3125	0.5034	0.5641	0.3481	0.5136	0.7092	0.5386	0.7086	0.2789	0.6316	0.2018	0.4563	0.2069	0.5123	0.2872	0.5116
20.3574	5.0	535	11.9065	0.3094	0.6001	0.2701	0.111	0.2743	0.4646	0.3037	0.4914	0.5537	0.3398	0.5075	0.7127	0.5381	0.6973	0.2998	0.6101	0.1896	0.4629	0.2168	0.5077	0.3028	0.4907
20.3574	6.0	642	11.7344	0.3247	0.6261	0.2856	0.1295	0.2678	0.511	0.3124	0.4959	0.5571	0.3649	0.4939	0.7194	0.5191	0.6851	0.3164	0.6013	0.2084	0.4808	0.2565	0.5138	0.3234	0.5044
20.3574	7.0	749	11.8164	0.3334	0.6308	0.3036	0.1224	0.2636	0.5347	0.3184	0.5033	0.5614	0.3559	0.5025	0.7285	0.5403	0.6932	0.3151	0.6278	0.213	0.471	0.2695	0.5062	0.3293	0.5089
20.3574	8.0	856	11.7534	0.3368	0.6435	0.3015	0.1196	0.273	0.5312	0.3174	0.4979	0.5609	0.3269	0.501	0.7413	0.558	0.7104	0.3457	0.6316	0.2289	0.4754	0.2288	0.4908	0.3223	0.4964
20.3574	9.0	963	11.8592	0.3326	0.6485	0.2827	0.1199	0.2747	0.5222	0.3153	0.501	0.5546	0.3213	0.4936	0.729	0.5518	0.6986	0.3311	0.6	0.2148	0.4839	0.236	0.4785	0.3295	0.512
12.6823	10.0	1070	11.9904	0.333	0.6405	0.2978	0.1135	0.2741	0.5196	0.3091	0.4947	0.5467	0.3229	0.4953	0.7097	0.5403	0.7018	0.3419	0.6215	0.2384	0.4688	0.2325	0.4615	0.3117	0.48
12.6823	11.0	1177	11.9472	0.334	0.6479	0.2904	0.1056	0.2845	0.5164	0.314	0.505	0.5605	0.3419	0.4978	0.7316	0.5548	0.7023	0.3488	0.6089	0.229	0.5027	0.2365	0.5	0.301	0.4889
12.6823	12.0	1284	12.0204	0.336	0.6377	0.3032	0.1228	0.2728	0.5224	0.3162	0.5077	0.5591	0.3479	0.5015	0.7214	0.5353	0.6896	0.3297	0.6025	0.2531	0.5031	0.252	0.5031	0.3099	0.4969
12.6823	13.0	1391	11.9460	0.3373	0.6461	0.2912	0.1245	0.2783	0.5243	0.3113	0.502	0.5521	0.3388	0.4936	0.7192	0.5445	0.6986	0.3387	0.6013	0.2517	0.4893	0.2403	0.4846	0.3114	0.4867
12.6823	14.0	1498	11.9051	0.3523	0.6617	0.3226	0.136	0.2944	0.5363	0.3223	0.5036	0.5558	0.3249	0.508	0.7127	0.57	0.709	0.3597	0.6076	0.2491	0.5049	0.2586	0.4662	0.3239	0.4916
11.1387	15.0	1605	12.0044	0.3399	0.6562	0.2971	0.1349	0.2898	0.5089	0.3074	0.5	0.556	0.3511	0.4995	0.7189	0.5576	0.7144	0.338	0.581	0.2422	0.4946	0.2619	0.4985	0.3	0.4916
11.1387	16.0	1712	12.2005	0.3263	0.6241	0.2928	0.1109	0.2512	0.5052	0.3029	0.4896	0.5467	0.3166	0.4913	0.704	0.5254	0.7005	0.3614	0.6101	0.2504	0.5049	0.2274	0.4477	0.267	0.4702
11.1387	17.0	1819	12.0237	0.3394	0.6504	0.3072	0.1154	0.2833	0.5157	0.3127	0.4958	0.5516	0.3447	0.5022	0.7017	0.5558	0.7009	0.353	0.6089	0.2552	0.4924	0.2226	0.4708	0.3103	0.4849
11.1387	18.0	1926	12.1186	0.3327	0.6432	0.2865	0.1113	0.2699	0.5206	0.3047	0.4767	0.5402	0.3143	0.4822	0.7094	0.5491	0.7063	0.3614	0.6076	0.2396	0.4799	0.2131	0.4462	0.3002	0.4609
10.0798	19.0	2033	12.0813	0.3357	0.6531	0.2953	0.1239	0.2789	0.5139	0.3065	0.4932	0.5437	0.337	0.4774	0.7108	0.5488	0.7068	0.3669	0.6127	0.2413	0.4661	0.23	0.4615	0.2914	0.4716
10.0798	20.0	2140	12.1951	0.3343	0.6419	0.3053	0.1129	0.2649	0.5183	0.3102	0.4806	0.5394	0.3237	0.4768	0.7028	0.5574	0.7104	0.3622	0.6013	0.2465	0.4768	0.2126	0.4338	0.2927	0.4747
10.0798	21.0	2247	12.2319	0.3353	0.6449	0.2919	0.111	0.2746	0.5182	0.3082	0.4866	0.5404	0.3251	0.4866	0.7057	0.5494	0.7117	0.3628	0.6051	0.2454	0.4777	0.2196	0.4308	0.2992	0.4769
10.0798	22.0	2354	12.2877	0.3294	0.6321	0.2919	0.1089	0.261	0.5186	0.3053	0.4851	0.5447	0.3214	0.4871	0.7145	0.5372	0.7077	0.3567	0.6316	0.2353	0.4723	0.2212	0.44	0.2967	0.472
10.0798	23.0	2461	12.1634	0.3319	0.6397	0.3058	0.1157	0.2597	0.5252	0.3122	0.4853	0.5394	0.3303	0.483	0.707	0.5459	0.7117	0.3409	0.5886	0.2364	0.4737	0.2308	0.4492	0.3055	0.4738
9.3418	24.0	2568	12.1011	0.3331	0.6406	0.2928	0.1156	0.2575	0.5196	0.3056	0.4807	0.539	0.3326	0.4804	0.7006	0.556	0.7176	0.3524	0.6038	0.2341	0.4603	0.2233	0.4477	0.2999	0.4658
9.3418	25.0	2675	12.3668	0.3304	0.6382	0.2966	0.1091	0.2597	0.5184	0.305	0.4758	0.5322	0.3251	0.4719	0.6994	0.5502	0.7104	0.3498	0.5949	0.2376	0.4643	0.2223	0.4262	0.2921	0.4653
9.3418	26.0	2782	12.1546	0.3373	0.6453	0.2976	0.1169	0.2658	0.5251	0.3092	0.4862	0.5425	0.3409	0.4803	0.7113	0.5548	0.7095	0.3569	0.6152	0.2476	0.4612	0.2262	0.4646	0.3009	0.4622
9.3418	27.0	2889	12.2520	0.3326	0.6367	0.2929	0.1101	0.2617	0.5187	0.304	0.4815	0.5366	0.3183	0.4799	0.7078	0.5496	0.709	0.3546	0.6076	0.2416	0.4594	0.2189	0.4492	0.2983	0.4578
9.3418	28.0	2996	12.2222	0.3312	0.6378	0.2931	0.1093	0.2587	0.5152	0.3034	0.4821	0.5379	0.3194	0.4758	0.7086	0.5549	0.7108	0.3513	0.6025	0.2382	0.4612	0.2156	0.4523	0.2962	0.4627
8.908	29.0	3103	12.2006	0.3346	0.6397	0.2931	0.1112	0.2634	0.5211	0.3047	0.4837	0.5396	0.329	0.4763	0.71	0.5522	0.7144	0.3559	0.6	0.2435	0.4647	0.2193	0.4523	0.302	0.4667
8.908	30.0	3210	12.2150	0.3305	0.6332	0.2924	0.1087	0.2627	0.5123	0.3018	0.4833	0.5384	0.3203	0.478	0.7067	0.5473	0.7158	0.3566	0.6063	0.2415	0.4621	0.2167	0.4431	0.2905	0.4649

Framework versions

• Transformers 4.44.0
• Pytorch 2.5.0+cu124
• Datasets 2.21.0
• Tokenizers 0.19.1