rtdetrv2-floorplan-v2

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: 6.2442
• Map: 0.6663
• Map 50: 0.758
• Map 75: 0.7005
• Map Small: 0.552
• Map Medium: 0.8059
• Map Large: 0.4146
• Mar 1: 0.274
• Mar 10: 0.8655
• Mar 100: 0.9038
• Mar Small: 0.7441
• Mar Medium: 0.9257
• Mar Large: 0.9488
• Map Window: 0.7714
• Mar 100 Window: 0.8784
• Map Door: 0.685
• Mar 100 Door: 0.9447
• Map Stair: 0.5424
• Mar 100 Stair: 0.8882

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: cosine
• lr_scheduler_warmup_steps: 0.05
• 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 Window	Mar 100 Window	Map Door	Mar 100 Door	Map Stair	Mar 100 Stair
13.6845	1.0	578	8.5528	0.5313	0.6844	0.5936	0.2287	0.5989	0.5458	0.2374	0.7215	0.7936	0.463	0.8541	0.7595	0.3682	0.7491	0.7475	0.8681	0.4782	0.7635
10.3385	2.0	1156	6.6908	0.5567	0.7148	0.6232	0.3899	0.6625	0.4741	0.2622	0.78	0.8265	0.5912	0.8732	0.8967	0.5611	0.768	0.6255	0.907	0.4834	0.8045
10.2151	3.0	1734	6.5601	0.5095	0.6472	0.567	0.3784	0.7081	0.4322	0.2631	0.802	0.8571	0.6519	0.9027	0.9034	0.5552	0.8032	0.6693	0.9175	0.3041	0.8507
9.2546	4.0	2312	6.2673	0.5949	0.7158	0.6618	0.404	0.7555	0.3886	0.2769	0.828	0.8686	0.6662	0.9078	0.9106	0.5946	0.8324	0.7551	0.9233	0.4351	0.8502
9.3453	5.0	2890	6.3085	0.6816	0.827	0.7517	0.4526	0.769	0.5305	0.282	0.8251	0.8727	0.6768	0.9113	0.9165	0.6705	0.8387	0.7792	0.9244	0.5951	0.855
8.4808	6.0	3468	5.9946	0.6344	0.7475	0.6899	0.4436	0.8077	0.502	0.2834	0.8366	0.8734	0.6808	0.9095	0.9277	0.6886	0.8405	0.7785	0.929	0.436	0.8505
9.0634	7.0	4046	6.1850	0.6574	0.7845	0.7219	0.4659	0.7887	0.4464	0.2847	0.8309	0.8664	0.682	0.8975	0.921	0.7034	0.8185	0.8197	0.9343	0.4491	0.8465
8.0622	8.0	4624	6.1631	0.6279	0.7624	0.6877	0.4289	0.7938	0.532	0.2853	0.8349	0.8749	0.6876	0.9137	0.9219	0.6222	0.8297	0.777	0.935	0.4846	0.86
7.6981	9.0	5202	6.0414	0.6407	0.7597	0.7031	0.4487	0.8103	0.4129	0.287	0.839	0.8853	0.7143	0.9198	0.9257	0.6327	0.845	0.7607	0.9359	0.5288	0.8751
8.3466	10.0	5780	6.1942	0.6848	0.8023	0.7421	0.4724	0.8127	0.5437	0.2883	0.8494	0.8895	0.7206	0.9219	0.936	0.6855	0.8541	0.8042	0.9382	0.5647	0.8761
7.5818	11.0	6358	5.9822	0.6707	0.772	0.7174	0.47	0.8111	0.4432	0.2881	0.8557	0.8907	0.7052	0.9259	0.9377	0.6817	0.859	0.7765	0.9424	0.5537	0.8706
7.2809	12.0	6936	6.1977	0.6297	0.7388	0.6821	0.5046	0.7594	0.3041	0.2698	0.8467	0.8892	0.7288	0.918	0.9238	0.6957	0.8572	0.6876	0.9427	0.5058	0.8678
7.6799	13.0	7514	6.1385	0.6041	0.7042	0.6452	0.4528	0.8024	0.2979	0.286	0.8481	0.8891	0.7027	0.9146	0.9471	0.5694	0.8608	0.7867	0.9401	0.4562	0.8663
7.2585	14.0	8092	6.1522	0.6609	0.7527	0.7107	0.5289	0.8068	0.4284	0.2889	0.8572	0.903	0.7482	0.9261	0.9359	0.7054	0.8739	0.727	0.943	0.5503	0.8922
7.0676	15.0	8670	6.2892	0.5739	0.6646	0.6126	0.5032	0.7629	0.2753	0.2741	0.848	0.8936	0.7282	0.9257	0.9307	0.6037	0.8608	0.6299	0.9432	0.4881	0.8766
7.1380	16.0	9248	6.1258	0.6293	0.7269	0.6734	0.5023	0.8094	0.4487	0.282	0.8548	0.8887	0.7237	0.9138	0.9288	0.7037	0.8527	0.6997	0.9427	0.4844	0.8706
7.1766	17.0	9826	6.1451	0.6694	0.769	0.7187	0.5442	0.8052	0.457	0.2847	0.8608	0.8984	0.7421	0.9267	0.9511	0.7256	0.8604	0.7286	0.9462	0.5541	0.8888
7.1360	18.0	10404	6.1668	0.6648	0.7519	0.712	0.566	0.7945	0.4634	0.2823	0.8622	0.9019	0.756	0.9291	0.9461	0.7593	0.8676	0.6829	0.9479	0.5523	0.8901
6.9559	19.0	10982	6.1588	0.6791	0.772	0.7246	0.557	0.8104	0.4372	0.2824	0.8611	0.8994	0.7424	0.9275	0.9527	0.759	0.864	0.7181	0.9462	0.5602	0.8881
7.1326	20.0	11560	6.1483	0.7176	0.8173	0.7628	0.5658	0.8179	0.4932	0.2821	0.8606	0.901	0.7299	0.9275	0.9474	0.7704	0.8707	0.7394	0.9471	0.6431	0.8853
7.0184	21.0	12138	6.2647	0.6733	0.7712	0.7192	0.5379	0.7962	0.4218	0.2772	0.8578	0.8983	0.7219	0.9292	0.9466	0.763	0.8685	0.6811	0.9424	0.5759	0.8841
6.6626	22.0	12716	6.1979	0.6668	0.7613	0.7041	0.531	0.8107	0.4587	0.2815	0.8613	0.8983	0.7323	0.9224	0.9371	0.7483	0.8653	0.6948	0.9437	0.5574	0.8858
6.7067	23.0	13294	6.1718	0.6794	0.7699	0.7209	0.5409	0.8136	0.459	0.2849	0.8658	0.9034	0.7467	0.9268	0.9445	0.7598	0.8752	0.7114	0.9444	0.5669	0.8905
6.4054	24.0	13872	6.2339	0.685	0.7732	0.7252	0.5606	0.8116	0.449	0.2848	0.8644	0.9013	0.7418	0.9256	0.9476	0.7725	0.8689	0.6926	0.9449	0.59	0.89
6.6368	25.0	14450	6.2354	0.6627	0.7539	0.6948	0.5542	0.8049	0.4134	0.2795	0.8661	0.9047	0.7434	0.9259	0.9464	0.7693	0.8788	0.6685	0.9432	0.5504	0.8922
6.4800	26.0	15028	6.2556	0.6755	0.7654	0.7069	0.5524	0.8075	0.4269	0.2829	0.865	0.9048	0.7387	0.9277	0.9464	0.7774	0.8784	0.6718	0.9447	0.5772	0.8913
6.2285	27.0	15606	6.2242	0.6806	0.7728	0.7109	0.5542	0.8111	0.3867	0.2777	0.8657	0.904	0.7425	0.9263	0.9472	0.7778	0.8779	0.702	0.9456	0.5621	0.8884
6.4170	28.0	16184	6.2169	0.6885	0.7807	0.7222	0.5544	0.8102	0.4346	0.2843	0.8687	0.9058	0.7461	0.9287	0.9468	0.7788	0.8802	0.7048	0.9456	0.5818	0.8917
6.4861	29.0	16762	6.2537	0.6699	0.7618	0.7051	0.552	0.8008	0.4108	0.2804	0.8642	0.9042	0.7428	0.9256	0.9481	0.7717	0.8761	0.6867	0.9453	0.5514	0.8912
6.2960	30.0	17340	6.2442	0.6663	0.758	0.7005	0.552	0.8059	0.4146	0.274	0.8655	0.9038	0.7441	0.9257	0.9488	0.7714	0.8784	0.685	0.9447	0.5424	0.8882

Framework versions

• Transformers 5.12.1
• Pytorch 2.12.1+cu130
• Datasets 5.0.0
• Tokenizers 0.22.2