You can deploy this model and start earning money today!

blzncz/
segformer-finetuned-4ss1st3r_s3gs3m_24Jan_rojo-10k-steps

Name: segformer-finetuned-4ss1st3r_s3gs3m_24Jan_rojo-10k-steps
Availability: InStock

generated_from_trainer

other

endpoints_compatible

Views

No views yet

segformer-finetuned-4ss1st3r_s3gs3m_24Jan_rojo-10k-steps

This model is a fine-tuned version of nvidia/mit-b0 on the blzncz/4ss1st3r_s3gs3m_24Jan_rojo dataset. It achieves the following results on the evaluation set:

Loss: 1.2098
Mean Iou: 0.3648
Mean Accuracy: 0.6821
Overall Accuracy: 0.6947
Accuracy Bg: nan
Accuracy Fallo cohesivo: 0.7354
Accuracy Fallo malla: 0.6052
Accuracy Fallo adhesivo: 0.9884
Accuracy Fallo burbuja: 0.3995
Iou Bg: 0.0
Iou Fallo cohesivo: 0.5920
Iou Fallo malla: 0.5774
Iou Fallo adhesivo: 0.2950
Iou Fallo burbuja: 0.3598

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: 6e-05
train_batch_size: 8
eval_batch_size: 8
seed: 1337
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: polynomial
training_steps: 10000

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Bg	Accuracy Fallo cohesivo	Accuracy Fallo malla	Accuracy Fallo adhesivo	Accuracy Fallo burbuja	Iou Fallo cohesivo	Iou Fallo malla	Iou Fallo adhesivo	Iou Fallo burbuja
0.5778	1.0	114	0.8590	0.2588	0.5626	0.6271	nan	0.3415	0.9120	0.9748	0.0221	0.3339	0.6211	0.3168	0.0220
0.3326	2.0	228	0.6845	0.3570	0.6755	0.7131	nan	0.5232	0.8953	0.9911	0.2921	0.5003	0.6924	0.3417	0.2503
0.2636	3.0	342	0.6662	0.3896	0.7107	0.7344	nan	0.7666	0.6622	0.9838	0.4304	0.6088	0.6188	0.4014	0.3191
0.2505	4.0	456	0.7666	0.3732	0.7408	0.6807	nan	0.4141	0.9417	0.9778	0.6297	0.4065	0.6276	0.4337	0.3980
0.2306	5.0	570	0.4680	0.4690	0.7389	0.8099	nan	0.7461	0.8649	0.9742	0.3705	0.6711	0.7095	0.6349	0.3294
0.1998	6.0	684	0.5711	0.4449	0.7494	0.7824	nan	0.8528	0.6732	0.9865	0.4850	0.6781	0.6338	0.5320	0.3807
0.2062	7.0	798	0.6403	0.4070	0.7437	0.7283	nan	0.5736	0.8683	0.9881	0.5447	0.5300	0.6452	0.4613	0.3987
0.182	8.0	912	0.5934	0.4344	0.7309	0.7770	nan	0.8171	0.7036	0.9840	0.4190	0.6640	0.6485	0.4916	0.3681
0.178	9.0	1026	0.7158	0.3811	0.6915	0.7313	nan	0.7292	0.6984	0.9913	0.3472	0.6148	0.6404	0.3348	0.3153
0.1568	10.0	1140	0.5892	0.4169	0.6970	0.7873	nan	0.8088	0.7398	0.9855	0.2538	0.6770	0.6664	0.5004	0.2407
0.1576	11.0	1254	0.6419	0.4228	0.7177	0.7652	nan	0.7970	0.7001	0.9805	0.3931	0.6509	0.6318	0.4701	0.3614
0.1667	12.0	1368	0.6563	0.4060	0.7369	0.7605	nan	0.7409	0.7517	0.9871	0.4681	0.6326	0.6731	0.4103	0.3139
0.1436	13.0	1482	0.9148	0.3864	0.7079	0.7187	nan	0.6666	0.7400	0.9900	0.4352	0.6025	0.6632	0.2829	0.3835
0.1469	14.0	1596	0.6680	0.4166	0.7216	0.7689	nan	0.7843	0.7225	0.9861	0.3936	0.6703	0.6608	0.3946	0.3571
0.1288	15.0	1710	0.8170	0.3765	0.6849	0.7164	nan	0.8269	0.5509	0.9859	0.3759	0.6242	0.5368	0.3815	0.3398
0.1243	16.0	1824	0.8197	0.4034	0.7169	0.7375	nan	0.8456	0.5776	0.9842	0.4602	0.6517	0.5582	0.4078	0.3991
0.1208	17.0	1938	0.7927	0.3848	0.6774	0.7295	nan	0.8592	0.5460	0.9810	0.3233	0.6359	0.5256	0.4647	0.2978
0.115	18.0	2052	1.1226	0.3376	0.6484	0.6727	nan	0.7053	0.5900	0.9905	0.3079	0.5659	0.5688	0.2673	0.2860
0.1138	19.0	2166	0.8244	0.4055	0.7099	0.7446	nan	0.8200	0.6248	0.9833	0.4115	0.6364	0.5964	0.4287	0.3659
0.1144	20.0	2280	0.5964	0.4493	0.7179	0.8034	nan	0.8594	0.7188	0.9808	0.3127	0.6995	0.6608	0.5990	0.2873
0.108	21.0	2394	0.6545	0.4418	0.7348	0.7902	nan	0.8263	0.7241	0.9835	0.4053	0.6832	0.6653	0.5023	0.3582
0.109	22.0	2508	0.9552	0.3775	0.6990	0.7058	nan	0.6835	0.6906	0.9894	0.4325	0.5907	0.6391	0.2756	0.3819
0.0987	23.0	2622	0.7971	0.3974	0.7133	0.7453	nan	0.7451	0.7124	0.9871	0.4084	0.6281	0.6560	0.3577	0.3452
0.0977	24.0	2736	0.9783	0.3718	0.6984	0.7001	nan	0.5950	0.7793	0.9916	0.4276	0.5491	0.6786	0.2620	0.3692
0.0954	25.0	2850	0.9562	0.3856	0.6981	0.7352	nan	0.7102	0.7294	0.9904	0.3623	0.6355	0.6603	0.2988	0.3332
0.0928	26.0	2964	0.9185	0.3787	0.6870	0.7355	nan	0.7815	0.6491	0.9847	0.3327	0.6569	0.6184	0.3151	0.3028
0.0918	27.0	3078	0.9617	0.3845	0.6916	0.7175	nan	0.8211	0.5605	0.9809	0.4037	0.6123	0.5462	0.3994	0.3648
0.0801	28.0	3192	1.1167	0.3672	0.6811	0.7091	nan	0.7352	0.6393	0.9927	0.3570	0.6151	0.6141	0.2816	0.3250
0.0852	29.0	3306	0.8549	0.4217	0.7108	0.7596	nan	0.8684	0.6040	0.9848	0.3862	0.6576	0.5808	0.5146	0.3553
0.0816	30.0	3420	0.9536	0.3902	0.7034	0.7366	nan	0.7752	0.6573	0.9885	0.3926	0.6415	0.6301	0.3274	0.3517
0.0876	31.0	3534	1.0597	0.3873	0.7065	0.7158	nan	0.7490	0.6374	0.9920	0.4475	0.6117	0.6160	0.3051	0.4035
0.0811	32.0	3648	0.8829	0.4038	0.7077	0.7569	nan	0.7949	0.6829	0.9860	0.3669	0.6442	0.6498	0.3943	0.3304
0.0789	33.0	3762	0.9615	0.4002	0.7104	0.7436	nan	0.7890	0.6575	0.9884	0.4066	0.6344	0.6308	0.3702	0.3658
0.0752	34.0	3876	0.7799	0.4297	0.7280	0.7806	nan	0.8279	0.6991	0.9873	0.3975	0.6787	0.6605	0.4458	0.3634
0.0731	35.0	3990	0.9285	0.4061	0.7025	0.7531	nan	0.8595	0.5987	0.9898	0.3619	0.6579	0.5797	0.4600	0.3330
0.0752	36.0	4104	0.9218	0.4112	0.7276	0.7463	nan	0.7632	0.6926	0.9880	0.4667	0.6393	0.6507	0.3462	0.4200
0.0701	37.0	4218	0.8808	0.4105	0.7184	0.7562	nan	0.8090	0.6635	0.9893	0.4119	0.6569	0.6342	0.3876	0.3740
0.0717	38.0	4332	1.1090	0.3748	0.6881	0.7166	nan	0.7554	0.6334	0.9905	0.3729	0.6272	0.6069	0.2969	0.3433
0.0716	39.0	4446	0.9456	0.4018	0.7064	0.7528	nan	0.8217	0.6418	0.9872	0.3747	0.6638	0.6131	0.3863	0.3456
0.069	40.0	4560	0.8462	0.4157	0.7038	0.7697	nan	0.8656	0.6316	0.9856	0.3324	0.6750	0.6041	0.4917	0.3078
0.07	41.0	4674	0.9715	0.3843	0.6886	0.7393	nan	0.8006	0.6353	0.9896	0.3289	0.6420	0.6104	0.3633	0.3056
0.0649	42.0	4788	0.9114	0.3997	0.7066	0.7592	nan	0.7682	0.7185	0.9917	0.3478	0.6613	0.6728	0.3449	0.3196
0.0665	43.0	4902	1.1847	0.3662	0.6812	0.6981	nan	0.7131	0.6389	0.9912	0.3817	0.5853	0.6122	0.2832	0.3504
0.0646	44.0	5016	1.1242	0.3744	0.6906	0.7086	nan	0.6930	0.6870	0.9891	0.3932	0.5902	0.6495	0.2770	0.3555
0.0662	45.0	5130	1.1017	0.3605	0.6735	0.7023	nan	0.7333	0.6261	0.9906	0.3439	0.5997	0.5996	0.2906	0.3126
0.0644	46.0	5244	1.2989	0.3470	0.6600	0.6735	nan	0.6567	0.6473	0.9917	0.3445	0.5607	0.6182	0.2377	0.3185
0.0595	47.0	5358	1.0764	0.3833	0.6982	0.7241	nan	0.7650	0.6389	0.9932	0.3957	0.6345	0.6134	0.3071	0.3618
0.0603	48.0	5472	1.0871	0.3692	0.6813	0.7128	nan	0.7153	0.6718	0.9884	0.3497	0.6079	0.6388	0.2797	0.3197
0.0591	49.0	5586	1.1054	0.3800	0.6956	0.7171	nan	0.7103	0.6866	0.9892	0.3963	0.6116	0.6458	0.2816	0.3609
0.0612	50.0	5700	1.1061	0.3652	0.6768	0.7087	nan	0.7394	0.6340	0.9903	0.3435	0.6027	0.6074	0.3009	0.3147
0.0609	51.0	5814	0.9938	0.3742	0.6850	0.7206	nan	0.7555	0.6433	0.9890	0.3523	0.6210	0.6121	0.3143	0.3235
0.058	52.0	5928	1.0391	0.3745	0.6836	0.7248	nan	0.7691	0.6374	0.9901	0.3379	0.6275	0.6082	0.3257	0.3109
0.0559	53.0	6042	0.9916	0.3922	0.7033	0.7373	nan	0.8044	0.6249	0.9902	0.3938	0.6429	0.6003	0.3644	0.3537
0.0572	54.0	6156	1.0124	0.3801	0.6907	0.7262	nan	0.7721	0.6371	0.9885	0.3650	0.6284	0.6052	0.3326	0.3341
0.0558	55.0	6270	1.0856	0.3692	0.6823	0.7120	nan	0.7232	0.6604	0.9894	0.3565	0.6094	0.6255	0.2864	0.3246
0.058	56.0	6384	1.0581	0.3837	0.6998	0.7212	nan	0.7353	0.6668	0.9910	0.4062	0.6126	0.6269	0.3125	0.3666
0.0518	57.0	6498	1.0176	0.3933	0.7060	0.7362	nan	0.7857	0.6440	0.9884	0.4060	0.6395	0.6127	0.3489	0.3655
0.0537	58.0	6612	1.2001	0.3676	0.6853	0.6947	nan	0.7737	0.5607	0.9884	0.4184	0.6003	0.5391	0.3221	0.3764
0.0552	59.0	6726	0.9751	0.3940	0.7068	0.7314	nan	0.8019	0.6139	0.9870	0.4244	0.6353	0.5871	0.3662	0.3816
0.0538	60.0	6840	1.0382	0.3782	0.6909	0.7203	nan	0.7216	0.6813	0.9895	0.3714	0.6093	0.6389	0.3011	0.3418
0.0528	61.0	6954	1.1785	0.3662	0.6819	0.7019	nan	0.7278	0.6310	0.9904	0.3784	0.5966	0.6013	0.2914	0.3419
0.0531	62.0	7068	1.1054	0.3685	0.6852	0.7026	nan	0.7290	0.6310	0.9899	0.3911	0.5969	0.5981	0.2961	0.3514
0.0522	63.0	7182	1.1271	0.3717	0.6871	0.7094	nan	0.7268	0.6496	0.9906	0.3816	0.6069	0.6148	0.2905	0.3460
0.0507	64.0	7296	1.0440	0.3734	0.6825	0.7242	nan	0.7678	0.6380	0.9884	0.3359	0.6279	0.6043	0.3272	0.3076
0.0519	65.0	7410	1.1191	0.3727	0.6884	0.7102	nan	0.7264	0.6517	0.9911	0.3843	0.6028	0.6156	0.2978	0.3472
0.0502	66.0	7524	1.0089	0.3917	0.7036	0.7408	nan	0.7555	0.6898	0.9896	0.3794	0.6413	0.6472	0.3261	0.3437
0.051	67.0	7638	1.2112	0.3672	0.6806	0.7083	nan	0.7352	0.6378	0.9899	0.3593	0.6078	0.6085	0.2918	0.3279
0.0508	68.0	7752	1.1584	0.3702	0.6860	0.7052	nan	0.7202	0.6477	0.9888	0.3875	0.5956	0.6155	0.2902	0.3495
0.048	69.0	7866	1.1363	0.3773	0.6922	0.7165	nan	0.7297	0.6628	0.9895	0.3865	0.6158	0.6289	0.2901	0.3518
0.0483	70.0	7980	1.1489	0.3749	0.6916	0.7103	nan	0.7398	0.6367	0.9889	0.4011	0.6074	0.6080	0.2994	0.3598
0.0495	71.0	8094	1.1470	0.3774	0.6943	0.7102	nan	0.7454	0.6295	0.9891	0.4131	0.6059	0.6032	0.3053	0.3724
0.0472	72.0	8208	1.2749	0.3597	0.6782	0.6864	nan	0.7291	0.5930	0.9891	0.4017	0.5899	0.5704	0.2771	0.3612
0.0486	73.0	8322	1.1217	0.3773	0.6946	0.7117	nan	0.7549	0.6224	0.9882	0.4128	0.6094	0.5946	0.3150	0.3678
0.051	74.0	8436	1.1895	0.3724	0.6889	0.7069	nan	0.7432	0.6247	0.9888	0.3990	0.6052	0.5959	0.3021	0.3590
0.0472	75.0	8550	1.2084	0.3677	0.6847	0.7009	nan	0.7179	0.6399	0.9905	0.3904	0.5979	0.6078	0.2808	0.3522
0.0481	76.0	8664	1.1778	0.3688	0.6841	0.7049	nan	0.7395	0.6244	0.9899	0.3824	0.6024	0.5950	0.2996	0.3469
0.0462	77.0	8778	1.2409	0.3693	0.6863	0.7015	nan	0.7278	0.6297	0.9900	0.3975	0.5964	0.5990	0.2918	0.3593
0.0464	78.0	8892	1.2724	0.3606	0.6792	0.6877	nan	0.7119	0.6158	0.9905	0.3986	0.5825	0.5857	0.2770	0.3578
0.0477	79.0	9006	1.2107	0.3629	0.6797	0.6936	nan	0.7322	0.6063	0.9898	0.3905	0.5928	0.5791	0.2889	0.3540
0.0452	80.0	9120	1.1745	0.3721	0.6889	0.7059	nan	0.7548	0.6087	0.9899	0.4022	0.6080	0.5820	0.3085	0.3620
0.0447	81.0	9234	1.2787	0.3599	0.6776	0.6876	nan	0.7199	0.6063	0.9902	0.3938	0.5857	0.5788	0.2786	0.3566
0.0481	82.0	9348	1.2049	0.3658	0.6836	0.6947	nan	0.7515	0.5865	0.9887	0.4078	0.5956	0.5627	0.3044	0.3660
0.0444	83.0	9462	1.1427	0.3746	0.6930	0.7051	nan	0.7520	0.6100	0.9883	0.4215	0.6042	0.5824	0.3100	0.3763
0.0481	84.0	9576	1.1876	0.3669	0.6848	0.6968	nan	0.7358	0.6094	0.9895	0.4046	0.5944	0.5818	0.2947	0.3636
0.046	85.0	9690	1.2264	0.3628	0.6799	0.6928	nan	0.7348	0.6015	0.9885	0.3948	0.5906	0.5746	0.2930	0.3560
0.0472	86.0	9804	1.2377	0.3659	0.6828	0.6967	nan	0.7287	0.6176	0.9890	0.3959	0.5926	0.5876	0.2913	0.3577
0.0465	87.0	9918	1.2037	0.3644	0.6841	0.6903	nan	0.7176	0.6150	0.9893	0.4146	0.5859	0.5856	0.2808	0.3697
0.0475	87.72	10000	1.2098	0.3648	0.6821	0.6947	nan	0.7354	0.6052	0.9884	0.3995	0.5920	0.5774	0.2950	0.3598

Framework versions

Transformers 4.31.0.dev0
Pytorch 2.0.1+cpu
Datasets 2.13.1
Tokenizers 0.13.3

blzncz/segformer-finetuned-4ss1st3r_s3gs3m_24Jan_rojo-10k-steps

segformer-finetuned-4ss1st3r_s3gs3m_24Jan_rojo-10k-steps

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

blzncz/
segformer-finetuned-4ss1st3r_s3gs3m_24Jan_rojo-10k-steps