rtdetr-v2-r50-final

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: 9.3098
Map: 0.484
Map 50: 0.8194
Map 75: 0.5685
Map Small: 0.4442
Map Medium: 0.619
Map Large: -1.0
Mar 1: 0.2974
Mar 10: 0.6181
Mar 100: 0.6181
Mar Small: 0.5923
Mar Medium: 0.6839
Mar Large: -1.0
Map Artemia: 0.484
Mar 100 Artemia: 0.6181

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: 300
num_epochs: 80

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 Artemia	Mar 100 Artemia
No log	1.0	250	18.2192	0.1176	0.2255	0.1067	0.1007	0.1663	-1.0	0.2536	0.5623	0.6062	0.557	0.6745	-1.0	0.1176	0.6062
303.3235	2.0	500	8.7055	0.4798	0.8537	0.4658	0.3973	0.5823	-1.0	0.3698	0.5897	0.6324	0.5903	0.6927	-1.0	0.4798	0.6324
303.3235	3.0	750	8.4912	0.3247	0.5772	0.3312	0.2884	0.5094	-1.0	0.3741	0.6009	0.6399	0.5962	0.7022	-1.0	0.3247	0.6399
14.0635	4.0	1000	8.1472	0.433	0.801	0.4254	0.4123	0.4874	-1.0	0.3788	0.5891	0.638	0.5995	0.6927	-1.0	0.433	0.638
14.0635	5.0	1250	8.5471	0.4823	0.8656	0.4874	0.3984	0.5908	-1.0	0.3785	0.6112	0.6218	0.5828	0.6759	-1.0	0.4823	0.6218
12.3863	6.0	1500	8.2223	0.2954	0.5425	0.2748	0.181	0.591	-1.0	0.3171	0.604	0.6308	0.5866	0.6927	-1.0	0.2954	0.6308
12.3863	7.0	1750	8.1330	0.4739	0.8798	0.4628	0.4009	0.5769	-1.0	0.3657	0.5972	0.61	0.5613	0.6781	-1.0	0.4739	0.61
11.5601	8.0	2000	8.6485	0.4427	0.8222	0.4373	0.3724	0.5556	-1.0	0.3639	0.5838	0.6106	0.5645	0.6752	-1.0	0.4427	0.6106
11.5601	9.0	2250	8.4338	0.4651	0.8454	0.4298	0.3886	0.5785	-1.0	0.3732	0.5885	0.6121	0.5677	0.6752	-1.0	0.4651	0.6121
10.6832	10.0	2500	8.7830	0.4116	0.7541	0.4007	0.3195	0.5685	-1.0	0.3364	0.5779	0.5913	0.543	0.6599	-1.0	0.4116	0.5913
10.6832	11.0	2750	9.5153	0.3823	0.7115	0.3608	0.2818	0.5922	-1.0	0.3355	0.5872	0.6047	0.5457	0.6876	-1.0	0.3823	0.6047
10.2254	12.0	3000	8.8212	0.4288	0.8069	0.3955	0.3362	0.5731	-1.0	0.3523	0.5679	0.576	0.5167	0.6584	-1.0	0.4288	0.576
10.2254	13.0	3250	9.4204	0.3342	0.6323	0.2957	0.2318	0.561	-1.0	0.3215	0.566	0.5791	0.5258	0.654	-1.0	0.3342	0.5791
9.6679	14.0	3500	9.0746	0.3941	0.739	0.3728	0.3044	0.5717	-1.0	0.3511	0.5657	0.5769	0.5258	0.6489	-1.0	0.3941	0.5769
9.6679	15.0	3750	8.9626	0.4389	0.8169	0.4097	0.3562	0.5757	-1.0	0.3583	0.5502	0.5526	0.4806	0.6526	-1.0	0.4389	0.5526
9.0927	16.0	4000	10.0535	0.3734	0.7551	0.2941	0.2795	0.5349	-1.0	0.3287	0.529	0.5324	0.4548	0.6401	-1.0	0.3734	0.5324
9.0927	17.0	4250	9.3608	0.4322	0.7876	0.4246	0.3399	0.5765	-1.0	0.3611	0.553	0.5561	0.4871	0.6518	-1.0	0.4322	0.5561
8.6476	18.0	4500	9.4866	0.4336	0.8246	0.3881	0.3505	0.5761	-1.0	0.3483	0.5548	0.5583	0.4919	0.6504	-1.0	0.4336	0.5583
8.6476	19.0	4750	8.9054	0.3521	0.6776	0.3219	0.2737	0.5382	-1.0	0.3442	0.5548	0.5555	0.486	0.6504	-1.0	0.3521	0.5555
8.4180	20.0	5000	10.3419	0.4022	0.7509	0.3569	0.3103	0.576	-1.0	0.3427	0.5464	0.5467	0.4774	0.6431	-1.0	0.4022	0.5467
8.4180	21.0	5250	9.3897	0.4097	0.7841	0.3714	0.321	0.5704	-1.0	0.3368	0.5433	0.5442	0.4731	0.6438	-1.0	0.4097	0.5442
7.9226	22.0	5500	9.6267	0.4193	0.7783	0.385	0.3328	0.5798	-1.0	0.3486	0.5424	0.5424	0.4704	0.6423	-1.0	0.4193	0.5424
7.9226	23.0	5750	9.8185	0.3414	0.6425	0.3189	0.2628	0.5344	-1.0	0.3333	0.5439	0.5449	0.4742	0.6445	-1.0	0.3414	0.5449
7.7066	24.0	6000	10.1001	0.3845	0.7315	0.3471	0.2966	0.5608	-1.0	0.3374	0.5436	0.5436	0.4806	0.6314	-1.0	0.3845	0.5436
7.7066	25.0	6250	10.0026	0.3333	0.6221	0.2944	0.2385	0.5714	-1.0	0.3006	0.5399	0.5399	0.4613	0.6489	-1.0	0.3333	0.5399
7.4507	26.0	6500	10.4097	0.387	0.7277	0.3349	0.2884	0.558	-1.0	0.3268	0.528	0.528	0.4505	0.6365	-1.0	0.387	0.528
7.4507	27.0	6750	11.5862	0.2979	0.576	0.2687	0.2081	0.5597	-1.0	0.2729	0.5305	0.5305	0.4608	0.6277	-1.0	0.2979	0.5305
7.2686	28.0	7000	11.1565	0.3373	0.6356	0.3059	0.2384	0.5668	-1.0	0.3012	0.5324	0.5324	0.4581	0.6358	-1.0	0.3373	0.5324
7.2686	29.0	7250	11.3274	0.3873	0.7167	0.353	0.298	0.5772	-1.0	0.3452	0.5489	0.5489	0.4823	0.6409	-1.0	0.3873	0.5489
7.0457	30.0	7500	10.8058	0.3777	0.7063	0.3456	0.2754	0.5825	-1.0	0.3174	0.5433	0.5433	0.4742	0.6394	-1.0	0.3777	0.5433
7.0457	31.0	7750	11.2777	0.2843	0.5369	0.2492	0.1902	0.5658	-1.0	0.2794	0.5399	0.5399	0.4672	0.6409	-1.0	0.2843	0.5399
6.7618	32.0	8000	10.6626	0.357	0.6757	0.328	0.2545	0.5687	-1.0	0.3215	0.5352	0.5352	0.4667	0.6307	-1.0	0.357	0.5352
6.7618	33.0	8250	11.5115	0.3621	0.6732	0.3219	0.2591	0.5758	-1.0	0.3255	0.5467	0.5467	0.4758	0.6453	-1.0	0.3621	0.5467
6.6498	34.0	8500	11.1793	0.2913	0.547	0.2776	0.1844	0.5777	-1.0	0.29	0.5355	0.5355	0.4586	0.6423	-1.0	0.2913	0.5355
6.6498	35.0	8750	11.4026	0.361	0.6539	0.343	0.2524	0.5849	-1.0	0.3109	0.5414	0.5414	0.4651	0.6467	-1.0	0.361	0.5414
6.4234	36.0	9000	11.2226	0.337	0.6255	0.3183	0.251	0.5519	-1.0	0.3193	0.5402	0.5402	0.4742	0.6321	-1.0	0.337	0.5402
6.4234	37.0	9250	12.2357	0.2855	0.5373	0.2639	0.1831	0.5486	-1.0	0.2785	0.5	0.5	0.4247	0.6044	-1.0	0.2855	0.5
6.3026	38.0	9500	12.0840	0.3422	0.6445	0.3124	0.2429	0.5701	-1.0	0.3162	0.5393	0.5393	0.4742	0.6299	-1.0	0.3422	0.5393
6.3026	39.0	9750	12.5118	0.3625	0.6741	0.3339	0.2529	0.5852	-1.0	0.315	0.5455	0.5455	0.472	0.6474	-1.0	0.3625	0.5455
6.1640	40.0	10000	10.9763	0.3497	0.6481	0.3373	0.2509	0.5721	-1.0	0.3202	0.5411	0.5411	0.4737	0.6343	-1.0	0.3497	0.5411
6.1640	41.0	10250	12.0927	0.3431	0.6341	0.3119	0.2359	0.5714	-1.0	0.315	0.5318	0.5318	0.457	0.6358	-1.0	0.3431	0.5318
5.9928	42.0	10500	11.7751	0.3721	0.6909	0.3479	0.2608	0.5858	-1.0	0.3315	0.5396	0.5396	0.4667	0.6401	-1.0	0.3721	0.5396
5.9928	43.0	10750	12.5522	0.3538	0.6562	0.3174	0.2437	0.5733	-1.0	0.3131	0.5368	0.5368	0.4656	0.6358	-1.0	0.3538	0.5368
5.8680	44.0	11000	12.3051	0.3586	0.6636	0.316	0.2448	0.5892	-1.0	0.3174	0.5411	0.5411	0.4586	0.6555	-1.0	0.3586	0.5411
5.8680	45.0	11250	12.0749	0.3623	0.6671	0.3333	0.2494	0.5861	-1.0	0.315	0.5492	0.5492	0.4731	0.6547	-1.0	0.3623	0.5492
5.7981	46.0	11500	11.2361	0.3645	0.6734	0.3494	0.2569	0.568	-1.0	0.3199	0.5312	0.5312	0.4629	0.6263	-1.0	0.3645	0.5312
5.7981	47.0	11750	13.0203	0.3411	0.6338	0.3156	0.2285	0.5893	-1.0	0.2966	0.5445	0.5445	0.4651	0.6547	-1.0	0.3411	0.5445
5.5465	48.0	12000	12.1057	0.2919	0.5339	0.276	0.1886	0.5287	-1.0	0.2611	0.4679	0.4679	0.3866	0.5818	-1.0	0.2919	0.4679
5.5465	49.0	12250	11.7633	0.367	0.6807	0.3299	0.2631	0.5751	-1.0	0.3237	0.5439	0.5439	0.4742	0.6401	-1.0	0.367	0.5439
5.4563	50.0	12500	12.3957	0.3423	0.6396	0.3162	0.2352	0.5742	-1.0	0.3084	0.5358	0.5358	0.4651	0.6343	-1.0	0.3423	0.5358
5.4563	51.0	12750	11.7921	0.3845	0.7063	0.3585	0.277	0.5873	-1.0	0.3368	0.5371	0.5371	0.4565	0.6489	-1.0	0.3845	0.5371
5.3358	52.0	13000	11.4805	0.3685	0.6793	0.3336	0.2555	0.5754	-1.0	0.3355	0.5368	0.5368	0.457	0.6474	-1.0	0.3685	0.5368
5.3358	53.0	13250	12.5466	0.3588	0.6529	0.3395	0.2435	0.5811	-1.0	0.3093	0.5321	0.5321	0.4505	0.6453	-1.0	0.3588	0.5321
5.2860	54.0	13500	13.0198	0.387	0.7045	0.3604	0.2713	0.5956	-1.0	0.3399	0.5511	0.5511	0.4715	0.6606	-1.0	0.387	0.5511
5.2860	55.0	13750	12.1157	0.3756	0.6887	0.3609	0.2705	0.5749	-1.0	0.3393	0.5408	0.5408	0.4704	0.6387	-1.0	0.3756	0.5408
5.0919	56.0	14000	12.7586	0.3681	0.6846	0.3341	0.2609	0.5777	-1.0	0.3336	0.5393	0.5393	0.4629	0.6453	-1.0	0.3681	0.5393
5.0919	57.0	14250	13.2146	0.35	0.6438	0.3284	0.2417	0.5711	-1.0	0.324	0.5346	0.5346	0.4581	0.6409	-1.0	0.35	0.5346
5.0850	58.0	14500	12.6233	0.3808	0.6939	0.35	0.2678	0.589	-1.0	0.3445	0.5458	0.5458	0.4672	0.6547	-1.0	0.3808	0.5458
5.0850	59.0	14750	13.0699	0.359	0.6616	0.3322	0.2502	0.5806	-1.0	0.3361	0.5433	0.5433	0.464	0.6533	-1.0	0.359	0.5433
4.9071	60.0	15000	13.1630	0.343	0.6327	0.3182	0.2361	0.5621	-1.0	0.3097	0.5131	0.5131	0.4296	0.6292	-1.0	0.343	0.5131
4.9071	61.0	15250	13.5266	0.3625	0.6662	0.34	0.2551	0.5722	-1.0	0.3221	0.5277	0.5277	0.4473	0.6394	-1.0	0.3625	0.5277
4.8413	62.0	15500	14.3104	0.3657	0.6698	0.3281	0.2611	0.5732	-1.0	0.3296	0.5383	0.5383	0.4656	0.6394	-1.0	0.3657	0.5383
4.8413	63.0	15750	13.2463	0.3598	0.6482	0.3504	0.2548	0.5667	-1.0	0.3218	0.5143	0.5143	0.4301	0.6314	-1.0	0.3598	0.5143
4.6595	64.0	16000	13.5959	0.3117	0.5716	0.2815	0.2112	0.5406	-1.0	0.2966	0.4857	0.4857	0.3995	0.6058	-1.0	0.3117	0.4857
4.6595	65.0	16250	13.6947	0.3739	0.6862	0.3443	0.2709	0.5751	-1.0	0.3315	0.5333	0.5333	0.4554	0.6416	-1.0	0.3739	0.5333
4.5445	66.0	16500	13.9606	0.3847	0.7065	0.3574	0.2767	0.5871	-1.0	0.3421	0.5461	0.5461	0.4677	0.6547	-1.0	0.3847	0.5461
4.5445	67.0	16750	13.8857	0.3656	0.6719	0.3467	0.2586	0.5773	-1.0	0.329	0.5327	0.5327	0.4516	0.6453	-1.0	0.3656	0.5327
4.4687	68.0	17000	13.2061	0.3873	0.7094	0.3566	0.2818	0.5793	-1.0	0.3399	0.5374	0.5374	0.4565	0.6496	-1.0	0.3873	0.5374
4.4687	69.0	17250	13.4241	0.3691	0.6811	0.3441	0.2669	0.5643	-1.0	0.329	0.524	0.524	0.4478	0.6299	-1.0	0.3691	0.524
4.3263	70.0	17500	13.8257	0.3612	0.6632	0.3292	0.2549	0.5761	-1.0	0.3265	0.5296	0.5296	0.4505	0.6394	-1.0	0.3612	0.5296
4.3263	71.0	17750	14.4978	0.3709	0.686	0.3507	0.2646	0.5773	-1.0	0.3358	0.538	0.538	0.4613	0.6445	-1.0	0.3709	0.538
4.3012	72.0	18000	14.1759	0.3702	0.6824	0.3478	0.2642	0.5707	-1.0	0.3315	0.5262	0.5262	0.4468	0.6365	-1.0	0.3702	0.5262
4.3012	73.0	18250	14.2601	0.3864	0.7112	0.3606	0.2804	0.5794	-1.0	0.3396	0.5383	0.5383	0.4608	0.646	-1.0	0.3864	0.5383
4.1892	74.0	18500	13.8906	0.3801	0.7004	0.3484	0.2789	0.572	-1.0	0.3371	0.5315	0.5315	0.4548	0.638	-1.0	0.3801	0.5315
4.1892	75.0	18750	14.6606	0.3701	0.6775	0.3358	0.2638	0.574	-1.0	0.3374	0.5346	0.5346	0.4586	0.6401	-1.0	0.3701	0.5346
4.1067	76.0	19000	14.3697	0.3804	0.7	0.3483	0.2746	0.5786	-1.0	0.3383	0.5374	0.5374	0.4602	0.6445	-1.0	0.3804	0.5374
4.1067	77.0	19250	14.4008	0.3757	0.6859	0.3534	0.2711	0.5771	-1.0	0.3393	0.5364	0.5364	0.4586	0.6445	-1.0	0.3757	0.5364
3.9826	78.0	19500	14.5908	0.3749	0.6871	0.3533	0.2666	0.5739	-1.0	0.3399	0.5333	0.5333	0.4559	0.6409	-1.0	0.3749	0.5333
3.9826	79.0	19750	14.8188	0.376	0.6855	0.354	0.2679	0.5795	-1.0	0.3386	0.5361	0.5361	0.4575	0.6453	-1.0	0.376	0.5361
3.9783	80.0	20000	14.8274	0.3774	0.6878	0.3564	0.2689	0.5803	-1.0	0.338	0.5371	0.5371	0.4575	0.6474	-1.0	0.3774	0.5371

Framework versions

Transformers 5.9.0
Pytorch 2.8.0+cu128
Datasets 4.2.0
Tokenizers 0.22.2

dariacuna/rtdetr-v2-r50-final

rtdetr-v2-r50-final

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种