rtdetr-v2-r101-cppe5-finetune-2

This model is a fine-tuned version of PekingU/rtdetr_v2_r101vd for the detection of coronary artery stenosis on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 9.5258
• Map: 0.3781
• Map 50: 0.9125
• Map 75: 0.2023
• Map Small: 0.2722
• Map Medium: 0.4358
• Map Large: 0.4439
• Mar 1: 0.4279
• Mar 10: 0.6136
• Mar 100: 0.6664
• Mar Small: 0.6158
• Mar Medium: 0.701
• Mar Large: 0.6333

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: 24
• 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: 50

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	313	18.7992	0.0102	0.037	0.0016	0.0193	0.023	0.0	0.0444	0.1863	0.2535	0.253	0.2555	0.0
112.163	2.0	626	14.0671	0.0806	0.265	0.0204	0.0711	0.0931	0.1998	0.1649	0.3478	0.4882	0.4476	0.5154	0.5667
112.163	3.0	939	13.0106	0.1549	0.5077	0.0356	0.1316	0.1788	0.2388	0.2393	0.4002	0.4952	0.4637	0.5172	0.4
12.4154	4.0	1252	12.8674	0.1861	0.578	0.0506	0.1487	0.2151	0.3809	0.2786	0.4329	0.5228	0.5074	0.5334	0.5
10.9002	5.0	1565	12.7759	0.1994	0.6122	0.0592	0.1562	0.2337	0.4435	0.2923	0.4511	0.5369	0.5092	0.5555	0.5667
10.9002	6.0	1878	12.9410	0.2477	0.6845	0.1046	0.1891	0.2878	0.4869	0.3295	0.4896	0.5643	0.5301	0.5872	0.6333
10.0597	7.0	2191	12.0453	0.2511	0.7021	0.092	0.1894	0.2916	0.4639	0.3144	0.4984	0.5683	0.5179	0.6024	0.6
9.4738	8.0	2504	11.6443	0.262	0.6911	0.1112	0.2077	0.3015	0.2426	0.3408	0.5317	0.5898	0.5452	0.6209	0.4667
9.4738	9.0	2817	11.4795	0.3091	0.7994	0.1487	0.2497	0.3488	0.3159	0.3764	0.5331	0.611	0.5664	0.6427	0.4
9.0887	10.0	3130	11.0460	0.3147	0.8058	0.1438	0.2394	0.365	0.469	0.3837	0.5331	0.5933	0.5301	0.6366	0.5333
9.0887	11.0	3443	10.8509	0.3038	0.7871	0.1337	0.214	0.3601	0.3414	0.3804	0.5429	0.5945	0.5435	0.6304	0.4
8.7187	12.0	3756	11.2568	0.2689	0.7366	0.1041	0.1987	0.3154	0.3722	0.3456	0.5245	0.5972	0.5402	0.6362	0.5667
8.5299	13.0	4069	10.6576	0.2818	0.7436	0.1189	0.2007	0.3285	0.2467	0.3433	0.5424	0.6164	0.5732	0.6466	0.5
8.5299	14.0	4382	10.3815	0.2688	0.6636	0.1296	0.1923	0.3198	0.395	0.3618	0.5586	0.6085	0.5518	0.6482	0.4333
8.2714	15.0	4695	10.2251	0.2846	0.7427	0.1324	0.1992	0.3356	0.3412	0.3408	0.5538	0.6187	0.5554	0.6623	0.5333
8.0708	16.0	5008	10.4468	0.2912	0.7585	0.1371	0.188	0.3503	0.2716	0.3442	0.5606	0.6252	0.5839	0.6551	0.3333
8.0708	17.0	5321	10.0924	0.3174	0.8149	0.1384	0.2167	0.3764	0.4464	0.3772	0.5462	0.6038	0.5426	0.6462	0.5
7.8789	18.0	5634	10.1511	0.292	0.7808	0.1069	0.1984	0.3441	0.4159	0.3466	0.5514	0.6103	0.5554	0.6484	0.5
7.8789	19.0	5947	10.0809	0.2872	0.7423	0.1304	0.1846	0.3477	0.3738	0.3546	0.558	0.6196	0.556	0.6638	0.4667
7.7461	20.0	6260	9.7884	0.3327	0.8237	0.1515	0.2486	0.3814	0.4109	0.396	0.579	0.6268	0.586	0.6555	0.4667
7.6557	21.0	6573	9.9407	0.3006	0.7914	0.1273	0.1849	0.3632	0.442	0.3455	0.5555	0.6118	0.5673	0.6429	0.4667
7.6557	22.0	6886	9.7508	0.3278	0.8365	0.1451	0.2254	0.3855	0.4052	0.372	0.5611	0.6179	0.567	0.6524	0.6333
7.4889	23.0	7199	9.9115	0.3163	0.8201	0.1201	0.2166	0.3695	0.4521	0.3826	0.566	0.6277	0.5869	0.6561	0.5333
7.4078	24.0	7512	9.9078	0.3068	0.7968	0.1297	0.1912	0.3724	0.5057	0.3533	0.5677	0.6291	0.5634	0.6743	0.5333
7.4078	25.0	7825	9.9138	0.2985	0.7844	0.1291	0.1905	0.3604	0.499	0.3514	0.5645	0.6385	0.5744	0.683	0.5
7.3325	26.0	8138	9.6711	0.3194	0.8042	0.1664	0.2146	0.3795	0.3995	0.3785	0.5969	0.6546	0.5905	0.6992	0.5
7.3325	27.0	8451	10.1850	0.2888	0.7886	0.1142	0.1822	0.3504	0.4257	0.3229	0.5677	0.6307	0.5616	0.6781	0.5667
7.2804	28.0	8764	10.1133	0.3185	0.8264	0.1482	0.2248	0.3728	0.4663	0.3688	0.5832	0.6509	0.5872	0.6953	0.4667
7.2029	29.0	9077	9.5707	0.3228	0.8097	0.1547	0.2271	0.3774	0.4653	0.3957	0.6044	0.6565	0.6071	0.6913	0.4667
7.2029	30.0	9390	10.1563	0.3377	0.8745	0.1421	0.2499	0.3885	0.3663	0.3789	0.5962	0.648	0.5988	0.683	0.4
7.0825	31.0	9703	9.8796	0.3364	0.8697	0.1503	0.242	0.3886	0.4629	0.3754	0.606	0.6556	0.6006	0.6939	0.5
7.026	32.0	10016	9.8638	0.345	0.8834	0.1628	0.2468	0.4006	0.4779	0.3856	0.6067	0.6615	0.6149	0.6937	0.5667
7.026	33.0	10329	9.6067	0.3561	0.8764	0.1902	0.2453	0.4162	0.399	0.3958	0.608	0.6585	0.6074	0.6947	0.4
6.8795	34.0	10642	9.5258	0.3781	0.9125	0.2023	0.2722	0.4358	0.4439	0.4279	0.6136	0.6664	0.6158	0.701	0.6333
6.8795	35.0	10955	9.4449	0.368	0.9007	0.188	0.2706	0.4204	0.5	0.4066	0.6088	0.6685	0.6253	0.699	0.5
6.7461	36.0	11268	9.2644	0.3658	0.8989	0.1945	0.2671	0.419	0.4653	0.4136	0.6142	0.6635	0.6202	0.6941	0.4667
6.64	37.0	11581	9.2426	0.3645	0.9017	0.1894	0.2627	0.4184	0.3975	0.4108	0.6078	0.6507	0.5997	0.6866	0.4333
6.64	38.0	11894	9.3995	0.3721	0.899	0.1976	0.2764	0.4243	0.4664	0.4192	0.6066	0.6544	0.6068	0.6877	0.5
6.4988	39.0	12207	9.4093	0.3769	0.9078	0.1964	0.2918	0.4253	0.4664	0.4288	0.6064	0.6531	0.6021	0.6887	0.5
6.4018	40.0	12520	9.3955	0.3615	0.8954	0.1759	0.2773	0.4089	0.4663	0.4116	0.6028	0.6546	0.5982	0.6941	0.4667
6.4018	41.0	12833	9.3966	0.3711	0.8973	0.1989	0.2758	0.4238	0.4993	0.421	0.6102	0.6635	0.6152	0.6968	0.6
6.3285	42.0	13146	9.4656	0.3655	0.8967	0.1968	0.2785	0.4172	0.4694	0.4245	0.6163	0.6701	0.6131	0.7097	0.5333
6.3285	43.0	13459	9.4390	0.3728	0.9149	0.1933	0.2914	0.419	0.4663	0.4295	0.6202	0.6675	0.6187	0.7018	0.4667
6.2397	44.0	13772	9.3682	0.3636	0.8795	0.202	0.2785	0.4146	0.4664	0.428	0.6148	0.6627	0.6065	0.7018	0.5
6.1381	45.0	14085	9.2842	0.3692	0.8998	0.2041	0.279	0.4205	0.4663	0.4279	0.6116	0.6625	0.6074	0.7012	0.4667
6.1381	46.0	14398	9.2333	0.3686	0.8987	0.2006	0.2758	0.4211	0.4663	0.428	0.6155	0.6621	0.6095	0.699	0.4667
6.0162	47.0	14711	9.2259	0.366	0.9051	0.1986	0.2756	0.4155	0.4663	0.4245	0.6157	0.6641	0.6128	0.7002	0.4667
5.9406	48.0	15024	9.2568	0.3632	0.902	0.1941	0.2729	0.4133	0.4733	0.4198	0.6145	0.6623	0.606	0.7014	0.5333
5.9406	49.0	15337	9.2850	0.3635	0.904	0.196	0.2733	0.4147	0.479	0.4223	0.616	0.6669	0.6134	0.704	0.5333
5.8549	50.0	15650	9.2647	0.3663	0.8999	0.1997	0.2746	0.4177	0.4663	0.4232	0.6149	0.664	0.6107	0.7014	0.4667

Framework versions

• Transformers 4.51.3
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1