license-plates-rtdetrv2

This model is a fine-tuned version of PekingU/rtdetr_v2_r18vd on the merve/license-plates dataset. It achieves the following results on the evaluation set:

• Loss: 4.6665
• Map: 0.5436
• Map 50: 0.8543
• Map 75: 0.6368
• Map Small: 0.3972
• Map Medium: 0.6773
• Map Large: 0.305
• Mar 1: 0.6232
• Mar 10: 0.7042
• Mar 100: 0.7389
• Mar Small: 0.5966
• Mar Medium: 0.7968
• Mar Large: 0.9
• Map License Plate: 0.5436
• Mar 100 License Plate: 0.7389

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_FUSED 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: 0.1
• num_epochs: 30.0
• 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 License Plate	Mar 100 License Plate
105.3917	1.0	78	64.5169	0.0003	0.0007	0.0002	0.0001	0.0007	0.0006	0.0	0.0105	0.1253	0.0345	0.1556	0.3667	0.0003	0.1253
28.5271	2.0	156	16.9015	0.1804	0.272	0.2199	0.0167	0.3254	0.0656	0.4674	0.5905	0.6547	0.3414	0.7937	0.7667	0.1804	0.6547
12.6303	3.0	234	6.8795	0.3897	0.6392	0.395	0.2355	0.5413	0.0254	0.5505	0.6747	0.6979	0.5103	0.7794	0.8	0.3897	0.6979
9.6897	4.0	312	5.5486	0.4711	0.7362	0.5617	0.3652	0.6532	0.0713	0.5821	0.6758	0.7168	0.5483	0.7905	0.8	0.4711	0.7168
8.6032	5.0	390	5.0634	0.5502	0.8263	0.6304	0.4174	0.6514	0.2223	0.6337	0.6979	0.7295	0.5897	0.7857	0.9	0.5502	0.7295
8.5343	6.0	468	4.8860	0.5714	0.8784	0.6796	0.4006	0.6749	0.449	0.6432	0.6895	0.7158	0.5759	0.7714	0.9	0.5714	0.7158
8.0221	7.0	546	4.8959	0.5302	0.8192	0.6206	0.4358	0.648	0.1917	0.6147	0.6937	0.7147	0.569	0.7714	0.9333	0.5302	0.7147
7.9080	8.0	624	4.7560	0.5744	0.8608	0.7203	0.4156	0.6822	0.2841	0.6379	0.7116	0.7274	0.5759	0.7873	0.9333	0.5744	0.7274
7.7790	9.0	702	4.8028	0.55	0.8335	0.712	0.4352	0.6779	0.2571	0.6432	0.7137	0.7421	0.6103	0.7937	0.9333	0.55	0.7421
7.8589	10.0	780	4.7129	0.5267	0.7739	0.6716	0.4461	0.7053	0.1132	0.5958	0.7063	0.7337	0.5793	0.7968	0.9	0.5267	0.7337
7.9087	11.0	858	4.6936	0.4752	0.7312	0.5722	0.419	0.6705	0.1515	0.6084	0.6747	0.7095	0.5621	0.7683	0.9	0.4752	0.7095
7.6768	12.0	936	4.6481	0.5802	0.857	0.7221	0.4442	0.7014	0.3169	0.6421	0.7116	0.7368	0.569	0.8048	0.9333	0.5802	0.7368
7.6139	13.0	1014	4.7139	0.5685	0.8597	0.7165	0.3979	0.6808	0.4735	0.6389	0.7	0.7253	0.5655	0.7905	0.9	0.5685	0.7253
7.5546	14.0	1092	4.6789	0.6017	0.9063	0.7801	0.4493	0.6661	0.8173	0.6526	0.7095	0.7253	0.5759	0.7857	0.9	0.6017	0.7253
7.3246	15.0	1170	4.6607	0.6005	0.9181	0.7523	0.4383	0.6712	0.8557	0.6547	0.6989	0.7274	0.5931	0.7794	0.9333	0.6005	0.7274
7.2602	16.0	1248	4.6885	0.5838	0.8985	0.6724	0.4336	0.6698	0.5133	0.6463	0.6905	0.7232	0.5724	0.7841	0.9	0.5838	0.7232
7.2412	17.0	1326	4.6910	0.5819	0.884	0.7182	0.4325	0.6971	0.4069	0.6526	0.7032	0.7358	0.5931	0.7937	0.9	0.5819	0.7358
7.3388	18.0	1404	4.6168	0.5955	0.9032	0.6925	0.4466	0.6858	0.5107	0.6579	0.7179	0.7474	0.6034	0.8063	0.9	0.5955	0.7474
7.1674	19.0	1482	4.6552	0.5854	0.8987	0.7051	0.434	0.6713	0.6445	0.6495	0.7211	0.7453	0.5897	0.8095	0.9	0.5854	0.7453
7.3368	20.0	1560	4.6603	0.5566	0.8663	0.6261	0.4406	0.6753	0.2842	0.6263	0.7137	0.7411	0.6	0.8	0.8667	0.5566	0.7411
7.0816	21.0	1638	4.6384	0.5738	0.8886	0.665	0.4351	0.6848	0.3486	0.6305	0.7074	0.7411	0.6069	0.7952	0.9	0.5738	0.7411
7.1841	22.0	1716	4.7157	0.562	0.8815	0.6775	0.4236	0.6646	0.4224	0.6347	0.7042	0.7421	0.6103	0.7952	0.9	0.562	0.7421
7.0787	23.0	1794	4.6758	0.5896	0.8971	0.6905	0.4232	0.684	0.6116	0.6358	0.7053	0.7484	0.6069	0.8048	0.9333	0.5896	0.7484
7.0592	24.0	1872	4.6918	0.5985	0.9165	0.6679	0.402	0.6843	0.817	0.6495	0.7074	0.7432	0.6138	0.7968	0.8667	0.5985	0.7432
6.8683	25.0	1950	4.6751	0.5493	0.8541	0.6423	0.419	0.6708	0.3486	0.6253	0.7053	0.7495	0.6034	0.8111	0.8667	0.5493	0.7495
6.8729	26.0	2028	4.7141	0.5554	0.8587	0.6308	0.4138	0.6729	0.3486	0.6368	0.7084	0.7421	0.6034	0.8	0.8667	0.5554	0.7421
7.1120	27.0	2106	4.6767	0.5829	0.8974	0.6834	0.3934	0.682	0.6115	0.6484	0.7063	0.7421	0.6	0.8	0.9	0.5829	0.7421
6.8692	28.0	2184	4.6875	0.5921	0.9078	0.7175	0.4151	0.6831	0.6114	0.6484	0.7084	0.7432	0.6034	0.8016	0.8667	0.5921	0.7432
6.9065	29.0	2262	4.6683	0.5705	0.8858	0.6659	0.4166	0.681	0.3865	0.6411	0.7032	0.7358	0.5931	0.7937	0.9	0.5705	0.7358
6.8172	30.0	2340	4.6665	0.5436	0.8543	0.6368	0.3972	0.6773	0.305	0.6232	0.7042	0.7389	0.5966	0.7968	0.9	0.5436	0.7389

Framework versions

• Transformers 5.3.0.dev0
• Pytorch 2.10.0+cu128
• Datasets 4.6.1
• Tokenizers 0.22.2