learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 60
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Mar 100	Map 50	Map 75
902.5145	1.0	71	335.3991	0.0	0.0319	0.0	0.0
340.8823	2.0	142	36.6198	0.0162	0.5894	0.0292	0.0164
28.8172	3.0	213	11.6351	0.2275	0.7191	0.4021	0.2194
18.9397	4.0	284	8.3844	0.5667	0.6908	0.9434	0.6165
14.3771	5.0	355	7.1939	0.5804	0.6972	0.9557	0.6253
13.7677	6.0	426	6.9469	0.5933	0.6894	0.9891	0.6531
13.7006	7.0	497	6.6969	0.592	0.7	0.9758	0.6705
13.2446	8.0	568	6.9132	0.5882	0.6851	0.9645	0.628
13.2452	9.0	639	6.7584	0.5963	0.7021	0.9711	0.6779
12.911	10.0	710	6.5749	0.5921	0.6887	0.9741	0.6289
12.9885	11.0	781	6.4069	0.6136	0.695	0.9782	0.7026
12.8467	12.0	852	6.4163	0.628	0.7035	0.9786	0.7375
12.6598	13.0	923	6.5697	0.6067	0.695	0.99	0.6764
12.4112	14.0	994	6.3900	0.6179	0.7043	0.9847	0.7073
12.6369	15.0	1065	6.4797	0.5973	0.6986	0.9839	0.6563
12.3894	16.0	1136	6.4122	0.6177	0.6957	0.9794	0.6821
12.4854	17.0	1207	6.3533	0.61	0.6972	0.9924	0.7234
12.3836	18.0	1278	6.3432	0.6148	0.6972	0.9902	0.6818
12.3067	19.0	1349	6.2016	0.6311	0.7135	0.9811	0.7726
12.4459	20.0	1420	6.4450	0.6183	0.7014	0.986	0.7104
12.2424	21.0	1491	6.3718	0.6209	0.7007	0.9886	0.7531
12.1976	22.0	1562	6.4085	0.6145	0.6936	0.9908	0.6676
11.8316	23.0	1633	6.1773	0.6341	0.7113	0.9895	0.765
12.1943	24.0	1704	6.2294	0.6331	0.7043	0.9912	0.7454
12.2331	25.0	1775	6.1884	0.6214	0.7071	0.9856	0.7324
11.7598	26.0	1846	6.2544	0.6337	0.7184	0.9811	0.7826
11.8849	27.0	1917	6.5367	0.6086	0.6773	0.9901	0.654
11.9108	28.0	1988	6.1798	0.6265	0.7128	0.9801	0.7468
11.7802	29.0	2059	6.4903	0.6002	0.6844	0.9854	0.6316
11.6292	30.0	2130	6.3403	0.6224	0.6922	0.9897	0.7199
11.6898	31.0	2201	6.2597	0.6264	0.7177	0.9798	0.7778
11.4146	32.0	2272	6.3491	0.6222	0.7014	0.9868	0.725
11.4249	33.0	2343	6.2945	0.6157	0.6908	0.9844	0.7287
11.4244	34.0	2414	6.3481	0.6106	0.6844	0.9884	0.7273
11.4353	35.0	2485	6.2424	0.631	0.6986	0.9895	0.7417
11.44	36.0	2556	6.4155	0.6127	0.6879	0.9788	0.6941
11.5167	37.0	2627	6.2776	0.627	0.695	0.9869	0.7384
11.3045	38.0	2698	6.3039	0.6184	0.6986	0.9814	0.7349
11.1152	39.0	2769	6.2876	0.6129	0.6972	0.9873	0.7217
11.1722	40.0	2840	6.3043	0.6208	0.7007	0.9797	0.7563
11.1581	41.0	2911	6.2575	0.626	0.7	0.9811	0.7226
11.3624	42.0	2982	6.2631	0.6203	0.7	0.9814	0.7321
11.0642	43.0	3053	6.2792	0.6217	0.7035	0.981	0.7281
11.1232	44.0	3124	6.3211	0.6198	0.6943	0.9844	0.7023
11.0043	45.0	3195	6.3222	0.6224	0.6929	0.9777	0.6996
10.8526	46.0	3266	6.3033	0.6279	0.6979	0.9861	0.7566
10.9216	47.0	3337	6.3102	0.6231	0.695	0.9842	0.7186
10.7362	48.0	3408	6.3306	0.6197	0.6957	0.9838	0.7148
10.9312	49.0	3479	6.2937	0.6228	0.6979	0.9823	0.7081
10.7711	50.0	3550	6.3390	0.6212	0.7028	0.9814	0.7207
10.7214	51.0	3621	6.3430	0.6187	0.6943	0.977	0.7095
10.6512	52.0	3692	6.3411	0.6195	0.7007	0.9753	0.7386
10.5377	53.0	3763	6.3898	0.6168	0.6908	0.9755	0.7073
10.6013	54.0	3834	6.3585	0.6159	0.6936	0.9757	0.6863
10.6208	55.0	3905	6.3540	0.6185	0.6979	0.9763	0.696
10.5331	56.0	3976	6.3733	0.619	0.6943	0.9761	0.693
10.8471	57.0	4047	6.3472	0.6195	0.695	0.9764	0.7064
10.6839	58.0	4118	6.3520	0.6163	0.6915	0.9755	0.7053
10.8091	59.0	4189	6.3618	0.6151	0.6908	0.9758	0.6847
10.4817	60.0	4260	6.3440	0.6179	0.6936	0.9767	0.7158

Framework versions

Transformers 4.57.1
Pytorch 2.9.1+cu128
Datasets 4.4.1
Tokenizers 0.22.1

ranm26/rtdetr-v2-r50-lpph-finetune-improved

rtdetr-v2-r50-lpph-finetune-improved

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

ranm26/
rtdetr-v2-r50-lpph-finetune-improved