yolo_finetuned_cards

This model is a fine-tuned version of hustvl/yolos-tiny on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5283
• Map: 0.4486
• Map 50: 0.5407
• Map 75: 0.4946
• Map Small: -1.0
• Map Medium: 0.3106
• Map Large: 0.5279
• Mar 1: 0.4769
• Mar 10: 0.7384
• Mar 100: 0.748
• Mar Small: -1.0
• Mar Medium: 0.6146
• Mar Large: 0.798
• Map Ace: 0.7527
• Mar 100 Ace: 0.8538
• Map Jack: 0.4254
• Mar 100 Jack: 0.875
• Map King: 0.3094
• Mar 100 King: 0.5391
• Map Nine: 0.4903
• Mar 100 Nine: 0.7846
• Map Queen: 0.3363
• Mar 100 Queen: 0.6077
• Map Ten: 0.3773
• Mar 100 Ten: 0.8278

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: 4
• 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
• 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 Ace	Mar 100 Ace	Map Jack	Mar 100 Jack	Map King	Mar 100 King	Map Nine	Mar 100 Nine	Map Queen	Mar 100 Queen	Map Ten	Mar 100 Ten
No log	1.0	74	1.8467	0.0072	0.0152	0.0069	-1.0	0.0093	0.0091	0.0359	0.1039	0.1384	-1.0	0.0562	0.1721	0.0001	0.0038	0.0086	0.0917	0.0141	0.4913	0.009	0.0769	0.0	0.0	0.0116	0.1667
No log	2.0	148	1.5411	0.015	0.0392	0.0083	-1.0	0.0182	0.0179	0.1051	0.2238	0.2546	-1.0	0.0924	0.3113	0.0147	0.2962	0.0247	0.2875	0.0162	0.487	0.0044	0.0846	0.023	0.15	0.0069	0.2222
No log	3.0	222	1.2203	0.0436	0.0849	0.0358	-1.0	0.083	0.0418	0.1913	0.324	0.3589	-1.0	0.2211	0.4047	0.0606	0.4692	0.0369	0.3375	0.0564	0.3957	0.0377	0.4154	0.0628	0.4192	0.0071	0.1167
No log	4.0	296	1.1673	0.0998	0.1735	0.0976	-1.0	0.1436	0.1039	0.2696	0.4261	0.4596	-1.0	0.4	0.478	0.181	0.6423	0.0732	0.4792	0.0579	0.4478	0.0977	0.4154	0.1326	0.5231	0.0562	0.25
No log	5.0	370	1.0189	0.105	0.1577	0.1138	-1.0	0.0941	0.1148	0.2639	0.3947	0.4346	-1.0	0.2178	0.5064	0.2451	0.6808	0.0408	0.3667	0.1266	0.7043	0.0906	0.4077	0.0774	0.2423	0.0495	0.2056
No log	6.0	444	0.9816	0.1211	0.1857	0.1403	-1.0	0.1133	0.1433	0.3085	0.4893	0.5198	-1.0	0.3164	0.5811	0.2268	0.6423	0.0627	0.4542	0.0832	0.4217	0.1366	0.7154	0.1308	0.5577	0.0864	0.3278
1.1436	7.0	518	1.0101	0.1435	0.2145	0.1886	-1.0	0.0975	0.1751	0.2949	0.4381	0.4543	-1.0	0.2683	0.5201	0.246	0.6692	0.0987	0.1708	0.1524	0.7435	0.2188	0.5462	0.0609	0.1962	0.0845	0.4
1.1436	8.0	592	0.7988	0.1474	0.1937	0.1744	-1.0	0.2069	0.1418	0.3337	0.4951	0.5194	-1.0	0.4162	0.5534	0.3347	0.8308	0.0433	0.2167	0.1334	0.7261	0.2306	0.6962	0.0051	0.0577	0.1372	0.5889
1.1436	9.0	666	0.7993	0.198	0.2647	0.2291	-1.0	0.2242	0.2115	0.3603	0.5675	0.6043	-1.0	0.5167	0.637	0.3802	0.8231	0.151	0.775	0.0963	0.5391	0.3429	0.7423	0.0555	0.1577	0.1618	0.5889
1.1436	10.0	740	0.7761	0.2328	0.3105	0.2895	-1.0	0.2438	0.2427	0.3857	0.5915	0.6072	-1.0	0.4681	0.6586	0.434	0.7962	0.2309	0.6583	0.0916	0.5087	0.3591	0.6885	0.1037	0.3192	0.1773	0.6722
1.1436	11.0	814	0.7794	0.22	0.3151	0.273	-1.0	0.1884	0.2471	0.3537	0.5626	0.5803	-1.0	0.4653	0.6195	0.4644	0.7731	0.1396	0.4458	0.1147	0.6913	0.3437	0.6769	0.0715	0.2115	0.1859	0.6833
1.1436	12.0	888	0.6468	0.3039	0.386	0.3483	-1.0	0.3153	0.3272	0.4105	0.6616	0.6842	-1.0	0.5194	0.7421	0.6112	0.8269	0.1726	0.725	0.1372	0.5739	0.4536	0.7192	0.2001	0.4269	0.2486	0.8333
1.1436	13.0	962	0.6429	0.3332	0.4269	0.3897	-1.0	0.2871	0.3685	0.4446	0.6414	0.6661	-1.0	0.5192	0.7133	0.612	0.8038	0.2212	0.7792	0.1907	0.4826	0.4713	0.7577	0.2201	0.3731	0.2839	0.8
0.6544	14.0	1036	0.6619	0.3203	0.4121	0.3606	-1.0	0.2484	0.3678	0.435	0.6089	0.6122	-1.0	0.4488	0.671	0.6788	0.8077	0.1918	0.4417	0.3214	0.7304	0.4129	0.7192	0.0912	0.2462	0.2255	0.7278
0.6544	15.0	1110	0.5787	0.3864	0.4838	0.4191	-1.0	0.3643	0.4212	0.4719	0.6883	0.702	-1.0	0.591	0.7415	0.6302	0.8615	0.2864	0.7625	0.3419	0.6652	0.4692	0.7308	0.1448	0.3308	0.4458	0.8611
0.6544	16.0	1184	0.5890	0.4038	0.5078	0.4519	-1.0	0.3721	0.4435	0.4556	0.686	0.7031	-1.0	0.588	0.7444	0.6156	0.85	0.2855	0.8458	0.3465	0.5	0.4875	0.7769	0.2168	0.3846	0.471	0.8611
0.6544	17.0	1258	0.6164	0.3537	0.4524	0.4078	-1.0	0.3051	0.4176	0.4632	0.6716	0.6771	-1.0	0.5484	0.7305	0.657	0.85	0.218	0.7125	0.3455	0.6696	0.4111	0.6038	0.1534	0.3769	0.337	0.85
0.6544	18.0	1332	0.5754	0.3857	0.4909	0.4301	-1.0	0.294	0.4424	0.4719	0.6662	0.6827	-1.0	0.5572	0.7314	0.723	0.8423	0.3596	0.8417	0.3559	0.5087	0.4101	0.7	0.2525	0.4423	0.2135	0.7611
0.6544	19.0	1406	0.6315	0.3812	0.4863	0.4301	-1.0	0.3352	0.4298	0.463	0.6618	0.6722	-1.0	0.5477	0.7239	0.6217	0.8077	0.3994	0.7792	0.2536	0.513	0.4066	0.6538	0.3022	0.4962	0.3037	0.7833
0.6544	20.0	1480	0.5835	0.3987	0.5095	0.4306	-1.0	0.3672	0.4446	0.4559	0.6985	0.7077	-1.0	0.6109	0.7412	0.6982	0.8346	0.3948	0.825	0.3084	0.5	0.4463	0.7769	0.2655	0.4654	0.279	0.8444
0.4562	21.0	1554	0.5678	0.428	0.5364	0.4814	-1.0	0.3144	0.4958	0.4689	0.6919	0.713	-1.0	0.5935	0.7567	0.7294	0.8654	0.4273	0.8625	0.2937	0.5174	0.4517	0.7154	0.2787	0.5231	0.3871	0.7944
0.4562	22.0	1628	0.5669	0.4063	0.4981	0.4623	-1.0	0.3407	0.469	0.4686	0.7198	0.7301	-1.0	0.5914	0.7811	0.6991	0.8731	0.4314	0.8833	0.3299	0.5739	0.4058	0.7231	0.2594	0.4769	0.3124	0.85
0.4562	23.0	1702	0.5492	0.4085	0.4977	0.4592	-1.0	0.3011	0.4752	0.4615	0.7106	0.7202	-1.0	0.5655	0.7787	0.7186	0.8538	0.4307	0.8708	0.3067	0.5391	0.4155	0.7038	0.3135	0.5423	0.2658	0.8111
0.4562	24.0	1776	0.5384	0.4196	0.5052	0.4682	-1.0	0.3428	0.4752	0.4706	0.7062	0.7219	-1.0	0.6051	0.7649	0.7412	0.8654	0.4333	0.875	0.3259	0.5391	0.4476	0.7423	0.2764	0.4654	0.293	0.8444
0.4562	25.0	1850	0.5365	0.4391	0.5315	0.4797	-1.0	0.3455	0.5024	0.476	0.7296	0.7404	-1.0	0.6231	0.7818	0.7436	0.8692	0.4165	0.8708	0.3093	0.5522	0.4575	0.7385	0.3527	0.5731	0.3551	0.8389
0.4562	26.0	1924	0.5333	0.442	0.5331	0.4914	-1.0	0.33	0.5148	0.4703	0.7241	0.7385	-1.0	0.5947	0.7906	0.7497	0.8538	0.4089	0.8708	0.3066	0.5304	0.4789	0.7615	0.339	0.5923	0.3691	0.8222
0.4562	27.0	1998	0.5292	0.4447	0.535	0.4941	-1.0	0.3092	0.5198	0.4558	0.7219	0.7315	-1.0	0.6164	0.7742	0.7453	0.85	0.4239	0.8667	0.3159	0.5348	0.4743	0.7538	0.3165	0.5615	0.3925	0.8222
0.3615	28.0	2072	0.5343	0.447	0.5419	0.4883	-1.0	0.3162	0.5262	0.4751	0.7289	0.7385	-1.0	0.6106	0.7845	0.7479	0.8538	0.4227	0.875	0.315	0.5435	0.4708	0.7538	0.3328	0.5769	0.3926	0.8278
0.3615	29.0	2146	0.5280	0.4485	0.5408	0.4885	-1.0	0.3106	0.5274	0.4753	0.7375	0.747	-1.0	0.6146	0.7967	0.7489	0.8538	0.4253	0.875	0.3094	0.5391	0.4882	0.7846	0.3337	0.5962	0.3854	0.8333
0.3615	30.0	2220	0.5283	0.4486	0.5407	0.4946	-1.0	0.3106	0.5279	0.4769	0.7384	0.748	-1.0	0.6146	0.798	0.7527	0.8538	0.4254	0.875	0.3094	0.5391	0.4903	0.7846	0.3363	0.6077	0.3773	0.8278

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2