segformer-b0_DsB1

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

Loss: 0.0111
Mean Iou: 0.7860
Mean Accuracy: 0.9662
Overall Accuracy: 0.9953
Accuracy Background: nan
Accuracy Haz: 0.9972
Accuracy Matrix: 0.9914
Accuracy Porosity: 0.8667
Accuracy Carbides: 0.9770
Accuracy Substrate: 0.9987
Iou Background: 0.0
Iou Haz: 0.9962
Iou Matrix: 0.9827
Iou Porosity: 0.7825
Iou Carbides: 0.9593
Iou Substrate: 0.9954

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: 4
eval_batch_size: 4
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Background	Accuracy Haz	Accuracy Matrix	Accuracy Porosity	Accuracy Carbides	Accuracy Substrate	Iou Haz	Iou Matrix	Iou Porosity	Iou Carbides	Iou Substrate
0.5096	1.0	103	0.2928	0.5652	0.7404	0.9287	nan	0.8944	0.9224	0.0	0.8967	0.9882	0.8848	0.8855	0.0	0.7567	0.8640
0.4342	2.0	206	0.3584	0.5372	0.7160	0.8652	nan	0.7273	0.9768	0.0	0.8787	0.9970	0.7241	0.9313	0.0	0.8454	0.7224
0.197	3.0	309	0.2423	0.5675	0.7370	0.9034	nan	0.8078	0.9829	0.0	0.8973	0.9970	0.8046	0.9431	0.0	0.8693	0.7879
0.1775	4.0	412	0.0727	0.6295	0.7750	0.9782	nan	0.9820	0.9714	0.0	0.9345	0.9871	0.9709	0.9480	0.0	0.8924	0.9657
0.1472	5.0	515	0.0645	0.6309	0.7783	0.9789	nan	0.9807	0.9676	0.0	0.9535	0.9896	0.9721	0.9487	0.0	0.8992	0.9656
0.1691	6.0	618	0.0679	0.6310	0.7761	0.9779	nan	0.9791	0.9835	0.0	0.9341	0.9838	0.9667	0.9579	0.0	0.9036	0.9577
0.0686	7.0	721	0.0527	0.6347	0.7798	0.9807	nan	0.9749	0.9866	0.0	0.9421	0.9952	0.9714	0.9605	0.0	0.9125	0.9639
1.2145	8.0	824	0.0417	0.6396	0.7824	0.9842	nan	0.9857	0.9841	0.0	0.9520	0.9904	0.9783	0.9651	0.0	0.9211	0.9730
0.0457	9.0	927	0.0430	0.6391	0.7859	0.9845	nan	0.9841	0.9724	0.0	0.9781	0.9948	0.9793	0.9623	0.0	0.9190	0.9739
2.0802	10.0	1030	0.0374	0.6412	0.7850	0.9853	nan	0.9953	0.9780	0.0	0.9723	0.9795	0.9799	0.9651	0.0	0.9280	0.9745
1.0817	11.0	1133	0.0345	0.6435	0.7862	0.9870	nan	0.9858	0.9835	0.0	0.9647	0.9968	0.9824	0.9684	0.0	0.9323	0.9777
0.2971	12.0	1236	0.0346	0.6438	0.7857	0.9873	nan	0.9902	0.9824	0.0	0.9633	0.9926	0.9835	0.9675	0.0	0.9324	0.9794
0.1245	13.0	1339	0.0292	0.6449	0.7863	0.9886	nan	0.9945	0.9799	0.0	0.9654	0.9916	0.9873	0.9660	0.0	0.9320	0.9842
0.0273	14.0	1442	0.0447	0.6413	0.7839	0.9833	nan	0.9961	0.9851	0.0005	0.9686	0.9690	0.9745	0.9708	0.0005	0.9367	0.9652
2.6071	15.0	1545	0.0267	0.6482	0.7891	0.9901	nan	0.9960	0.9873	0.0050	0.9676	0.9897	0.9880	0.9730	0.0050	0.9388	0.9845
0.0195	16.0	1648	0.0231	0.6495	0.7904	0.9909	nan	0.9935	0.9889	0.0076	0.9669	0.9951	0.9899	0.9725	0.0076	0.9395	0.9873
0.348	17.0	1751	0.0224	0.6618	0.8041	0.9907	nan	0.9951	0.9887	0.0828	0.9603	0.9933	0.9896	0.9722	0.0828	0.9389	0.9871
0.1579	18.0	1854	0.0267	0.6733	0.8201	0.9893	nan	0.9878	0.9861	0.1576	0.9707	0.9983	0.9862	0.9732	0.1576	0.9405	0.9821
0.0084	19.0	1957	0.0235	0.6741	0.8215	0.9904	nan	0.9919	0.9827	0.1583	0.9778	0.9966	0.9893	0.9722	0.1583	0.9380	0.9869
2.5181	20.0	2060	0.0218	0.7351	0.8938	0.9911	nan	0.9965	0.9840	0.5229	0.9740	0.9916	0.9900	0.9735	0.5189	0.9413	0.9871
0.0315	21.0	2163	0.0265	0.7395	0.8996	0.9897	nan	0.9872	0.9879	0.5534	0.9711	0.9987	0.9863	0.9756	0.5502	0.9431	0.9820
0.0392	22.0	2266	0.0208	0.7549	0.9186	0.9917	nan	0.9937	0.9895	0.6512	0.9624	0.9964	0.9909	0.9751	0.6336	0.9415	0.9884
0.0591	23.0	2369	0.0219	0.7572	0.9256	0.9913	nan	0.9927	0.9822	0.6759	0.9800	0.9970	0.9904	0.9740	0.6496	0.9414	0.9876
0.0169	24.0	2472	0.0212	0.7647	0.9361	0.9917	nan	0.9952	0.9856	0.7320	0.9737	0.9943	0.9909	0.9753	0.6902	0.9436	0.9882
0.0215	25.0	2575	0.0212	0.7657	0.9367	0.9919	nan	0.9937	0.9886	0.7371	0.9676	0.9965	0.9908	0.9761	0.6933	0.9457	0.9883
0.0324	26.0	2678	0.0180	0.7630	0.9327	0.9926	nan	0.9964	0.9869	0.7100	0.9756	0.9944	0.9924	0.9766	0.6730	0.9455	0.9903
0.1575	27.0	2781	0.0200	0.7714	0.9505	0.9921	nan	0.9958	0.9852	0.8043	0.9721	0.9951	0.9912	0.9751	0.7262	0.9463	0.9894
0.7673	28.0	2884	0.0185	0.7716	0.9523	0.9929	nan	0.9969	0.9876	0.8140	0.9677	0.9954	0.9928	0.9759	0.7233	0.9461	0.9913
0.0199	29.0	2987	0.0192	0.7696	0.9475	0.9922	nan	0.9951	0.9874	0.7881	0.9718	0.9952	0.9913	0.9768	0.7122	0.9477	0.9897
0.016	30.0	3090	0.0197	0.7702	0.9491	0.9920	nan	0.9924	0.9859	0.7933	0.9755	0.9983	0.9912	0.9764	0.7192	0.9454	0.9890
0.0053	31.0	3193	0.0180	0.7613	0.9270	0.9926	nan	0.9968	0.9924	0.6927	0.9593	0.9939	0.9925	0.9760	0.6650	0.9439	0.9906
0.0241	32.0	3296	0.0176	0.7704	0.9441	0.9928	nan	0.9969	0.9908	0.7759	0.9624	0.9946	0.9929	0.9766	0.7157	0.9458	0.9911
0.0265	33.0	3399	0.0186	0.7720	0.9509	0.9920	nan	0.9979	0.9919	0.8137	0.9607	0.9902	0.9909	0.9771	0.7297	0.9460	0.9880
0.0771	34.0	3502	0.0165	0.7717	0.9443	0.9933	nan	0.9967	0.9884	0.7682	0.9725	0.9959	0.9935	0.9778	0.7195	0.9476	0.9917
0.2239	35.0	3605	0.0172	0.7739	0.9521	0.9928	nan	0.9943	0.9886	0.8083	0.9716	0.9976	0.9923	0.9777	0.7340	0.9488	0.9905
0.0336	36.0	3708	0.0165	0.7721	0.9437	0.9932	nan	0.9969	0.9894	0.7690	0.9674	0.9956	0.9932	0.9771	0.7225	0.9479	0.9916
0.3056	37.0	3811	0.0159	0.7741	0.9488	0.9937	nan	0.9967	0.9858	0.7843	0.9803	0.9968	0.9941	0.9780	0.7314	0.9484	0.9925
0.0129	38.0	3914	0.0180	0.7736	0.9473	0.9933	nan	0.9960	0.9896	0.7852	0.9693	0.9968	0.9934	0.9778	0.7294	0.9491	0.9918
0.0197	39.0	4017	0.0162	0.7757	0.9578	0.9937	nan	0.9975	0.9884	0.8346	0.9729	0.9956	0.9941	0.9782	0.7391	0.9501	0.9926
0.0154	40.0	4120	0.0165	0.7741	0.9474	0.9937	nan	0.9970	0.9902	0.7825	0.9714	0.9959	0.9939	0.9788	0.7291	0.9502	0.9924
0.0609	41.0	4223	0.0170	0.7750	0.9532	0.9937	nan	0.9966	0.9902	0.8136	0.9687	0.9967	0.9939	0.9784	0.7353	0.9500	0.9924
0.0089	42.0	4326	0.0158	0.7764	0.9523	0.9939	nan	0.9977	0.9902	0.8070	0.9712	0.9953	0.9940	0.9791	0.7409	0.9516	0.9925
0.0075	43.0	4429	0.0266	0.7747	0.9568	0.9915	nan	0.9921	0.9885	0.8324	0.9753	0.9956	0.9889	0.9790	0.7442	0.9509	0.9853
0.0178	44.0	4532	0.0147	0.7735	0.9443	0.9939	nan	0.9970	0.9895	0.7668	0.9715	0.9968	0.9944	0.9785	0.7246	0.9505	0.9931
0.0458	45.0	4635	0.0149	0.7766	0.9512	0.9941	nan	0.9979	0.9885	0.7974	0.9765	0.9959	0.9945	0.9793	0.7410	0.9517	0.9932
0.0099	46.0	4738	0.0139	0.7755	0.9481	0.9942	nan	0.9968	0.9888	0.7821	0.9754	0.9976	0.9947	0.9792	0.7337	0.9518	0.9935
0.0129	47.0	4841	0.0139	0.7783	0.9568	0.9943	nan	0.9964	0.9880	0.8236	0.9776	0.9984	0.9949	0.9794	0.7495	0.9525	0.9938
0.0293	48.0	4944	0.0148	0.7780	0.9533	0.9941	nan	0.9968	0.9876	0.8078	0.9766	0.9977	0.9945	0.9787	0.7490	0.9523	0.9935
0.1331	49.0	5047	0.0141	0.7782	0.9562	0.9940	nan	0.9972	0.9908	0.8252	0.9718	0.9961	0.9943	0.9799	0.7497	0.9522	0.9929
0.049	50.0	5150	0.0157	0.7794	0.9594	0.9941	nan	0.9955	0.9900	0.8391	0.9736	0.9988	0.9942	0.9799	0.7562	0.9530	0.9929
0.0344	51.0	5253	0.0153	0.7796	0.9610	0.9943	nan	0.9975	0.9876	0.8443	0.9789	0.9968	0.9946	0.9796	0.7565	0.9535	0.9936
0.0389	52.0	5356	0.0174	0.7778	0.9571	0.9931	nan	0.9941	0.9924	0.8330	0.9682	0.9975	0.9922	0.9802	0.7527	0.9519	0.9899
0.0011	53.0	5459	0.0150	0.7777	0.9495	0.9943	nan	0.9963	0.9883	0.7862	0.9788	0.9981	0.9949	0.9798	0.7451	0.9527	0.9936
0.0147	54.0	5562	0.0138	0.7790	0.9571	0.9940	nan	0.9956	0.9870	0.8230	0.9811	0.9987	0.9945	0.9796	0.7544	0.9526	0.9930
0.0267	55.0	5665	0.0127	0.7808	0.9623	0.9944	nan	0.9967	0.9869	0.8476	0.9823	0.9981	0.9953	0.9798	0.7626	0.9530	0.9942
0.0965	56.0	5768	0.0133	0.7762	0.9456	0.9945	nan	0.9979	0.9911	0.7703	0.9723	0.9965	0.9953	0.9802	0.7347	0.9529	0.9941
0.0251	57.0	5871	0.0141	0.7799	0.9593	0.9942	nan	0.9963	0.9867	0.8344	0.9804	0.9985	0.9951	0.9790	0.7589	0.9522	0.9939
0.0048	58.0	5974	0.0129	0.7810	0.9608	0.9947	nan	0.9972	0.9895	0.8422	0.9776	0.9977	0.9955	0.9806	0.7615	0.9543	0.9944
0.0172	59.0	6077	0.0138	0.7794	0.9554	0.9942	nan	0.9956	0.9897	0.8178	0.9749	0.9989	0.9947	0.9801	0.7543	0.9540	0.9933
0.0605	60.0	6180	0.0136	0.7802	0.9567	0.9943	nan	0.9955	0.9899	0.8225	0.9766	0.9989	0.9946	0.9807	0.7591	0.9539	0.9932
1.8879	61.0	6283	0.0124	0.7809	0.9547	0.9948	nan	0.9973	0.9912	0.8133	0.9739	0.9977	0.9956	0.9811	0.7596	0.9546	0.9943
0.0341	62.0	6386	0.0120	0.7807	0.9567	0.9947	nan	0.9977	0.9879	0.8197	0.9807	0.9972	0.9956	0.9803	0.7598	0.9537	0.9946
0.0201	63.0	6489	0.0120	0.7818	0.9620	0.9948	nan	0.9973	0.9896	0.8489	0.9761	0.9980	0.9957	0.9805	0.7655	0.9545	0.9948
0.0168	64.0	6592	0.0118	0.7820	0.9664	0.9949	nan	0.9979	0.9914	0.8721	0.9733	0.9972	0.9957	0.9813	0.7648	0.9554	0.9947
0.0038	65.0	6695	0.0117	0.7808	0.9565	0.9947	nan	0.9975	0.9938	0.8288	0.9649	0.9976	0.9958	0.9803	0.7612	0.9528	0.9947
0.0481	66.0	6798	0.0117	0.7820	0.9561	0.9950	nan	0.9969	0.9911	0.8183	0.9758	0.9985	0.9957	0.9815	0.7642	0.9556	0.9947
0.0134	67.0	6901	0.0122	0.7832	0.9620	0.9949	nan	0.9976	0.9900	0.8480	0.9769	0.9976	0.9956	0.9811	0.7714	0.9561	0.9946
0.0006	68.0	7004	0.0118	0.7821	0.9615	0.9949	nan	0.9977	0.9902	0.8467	0.9755	0.9976	0.9957	0.9809	0.7658	0.9558	0.9947
0.0237	69.0	7107	0.0118	0.7828	0.9581	0.9950	nan	0.9975	0.9879	0.8245	0.9824	0.9982	0.9960	0.9809	0.7689	0.9556	0.9951
0.0061	70.0	7210	0.0123	0.7830	0.9619	0.9948	nan	0.9969	0.9910	0.8481	0.9756	0.9981	0.9954	0.9816	0.7702	0.9569	0.9942
0.0214	71.0	7313	0.0131	0.7832	0.9601	0.9948	nan	0.9963	0.9919	0.8408	0.9730	0.9988	0.9954	0.9814	0.7724	0.9560	0.9943
0.0085	72.0	7416	0.0120	0.7840	0.9682	0.9950	nan	0.9972	0.9910	0.8789	0.9757	0.9980	0.9957	0.9818	0.7750	0.9569	0.9947
0.0048	73.0	7519	0.0125	0.7827	0.9626	0.9949	nan	0.9962	0.9898	0.8490	0.9791	0.9991	0.9955	0.9817	0.7678	0.9567	0.9944
0.0033	74.0	7622	0.0127	0.7827	0.9576	0.9950	nan	0.9972	0.9919	0.8262	0.9747	0.9982	0.9957	0.9820	0.7667	0.9568	0.9947
0.0044	75.0	7725	0.0117	0.7812	0.9513	0.9952	nan	0.9970	0.9912	0.7913	0.9782	0.9985	0.9958	0.9824	0.7572	0.9574	0.9947
0.0041	76.0	7828	0.0118	0.7844	0.9624	0.9951	nan	0.9975	0.9909	0.8486	0.9770	0.9981	0.9960	0.9820	0.7759	0.9575	0.9950
0.0091	77.0	7931	0.0113	0.7839	0.9624	0.9951	nan	0.9979	0.9901	0.8477	0.9788	0.9976	0.9960	0.9818	0.7731	0.9573	0.9951
0.0181	78.0	8034	0.0121	0.7844	0.9636	0.9952	nan	0.9971	0.9916	0.8552	0.9755	0.9986	0.9960	0.9822	0.7754	0.9578	0.9949
0.0169	79.0	8137	0.0110	0.7850	0.9644	0.9953	nan	0.9977	0.9914	0.8589	0.9755	0.9983	0.9962	0.9821	0.7780	0.9579	0.9954
0.0136	80.0	8240	0.0111	0.7832	0.9573	0.9952	nan	0.9971	0.9904	0.8210	0.9793	0.9987	0.9961	0.9821	0.7687	0.9575	0.9951
0.0098	81.0	8343	0.0104	0.7847	0.9614	0.9953	nan	0.9978	0.9904	0.8420	0.9787	0.9980	0.9963	0.9822	0.7760	0.9581	0.9955
0.0073	82.0	8446	0.0109	0.7839	0.9584	0.9954	nan	0.9978	0.9900	0.8257	0.9803	0.9982	0.9964	0.9821	0.7716	0.9577	0.9956
0.0298	83.0	8549	0.0115	0.7848	0.9633	0.9951	nan	0.9966	0.9903	0.8513	0.9790	0.9991	0.9958	0.9823	0.7784	0.9579	0.9947
0.0237	84.0	8652	0.0117	0.7851	0.9631	0.9952	nan	0.9967	0.9922	0.8522	0.9753	0.9991	0.9959	0.9826	0.7792	0.9583	0.9949
0.0047	85.0	8755	0.0123	0.7856	0.9668	0.9951	nan	0.9965	0.9912	0.8697	0.9777	0.9990	0.9957	0.9826	0.7820	0.9589	0.9947
0.0398	86.0	8858	0.0123	0.7859	0.9705	0.9951	nan	0.9967	0.9902	0.8880	0.9788	0.9990	0.9957	0.9822	0.7838	0.9590	0.9947
0.0158	87.0	8961	0.0119	0.7853	0.9611	0.9954	nan	0.9972	0.9922	0.8408	0.9766	0.9986	0.9961	0.9829	0.7790	0.9589	0.9952
0.0072	88.0	9064	0.0116	0.7858	0.9656	0.9952	nan	0.9968	0.9905	0.8622	0.9795	0.9990	0.9958	0.9826	0.7825	0.9590	0.9947
0.0135	89.0	9167	0.0117	0.7861	0.9655	0.9954	nan	0.9975	0.9896	0.8601	0.9820	0.9984	0.9962	0.9825	0.7840	0.9585	0.9954
0.0036	90.0	9270	0.0111	0.7856	0.9633	0.9953	nan	0.9976	0.9911	0.8515	0.9780	0.9981	0.9962	0.9826	0.7805	0.9588	0.9953
0.0098	91.0	9373	0.0110	0.7855	0.9631	0.9953	nan	0.9970	0.9912	0.8506	0.9781	0.9988	0.9960	0.9827	0.7801	0.9589	0.9951
0.0106	92.0	9476	0.0106	0.7853	0.9640	0.9954	nan	0.9973	0.9917	0.8555	0.9769	0.9988	0.9962	0.9828	0.7784	0.9591	0.9953
0.0048	93.0	9579	0.0106	0.7856	0.9665	0.9954	nan	0.9976	0.9902	0.8656	0.9810	0.9982	0.9963	0.9827	0.7801	0.9591	0.9954
0.0104	94.0	9682	0.0106	0.7852	0.9616	0.9955	nan	0.9977	0.9910	0.8411	0.9798	0.9984	0.9964	0.9829	0.7772	0.9592	0.9956
0.0249	95.0	9785	0.0108	0.7858	0.9647	0.9954	nan	0.9973	0.9910	0.8574	0.9792	0.9988	0.9962	0.9829	0.7813	0.9593	0.9953
0.006	96.0	9888	0.0109	0.7853	0.9620	0.9955	nan	0.9971	0.9911	0.8427	0.9800	0.9989	0.9963	0.9830	0.7777	0.9593	0.9953
0.0385	97.0	9991	0.0118	0.7857	0.9611	0.9953	nan	0.9966	0.9913	0.8393	0.9789	0.9991	0.9959	0.9830	0.7809	0.9594	0.9949
0.0034	98.0	10094	0.0109	0.7859	0.9618	0.9955	nan	0.9974	0.9911	0.8428	0.9790	0.9987	0.9963	0.9829	0.7815	0.9594	0.9954
0.0018	99.0	10197	0.0111	0.7865	0.9659	0.9955	nan	0.9973	0.9907	0.8627	0.9798	0.9988	0.9962	0.9829	0.7848	0.9595	0.9954
0.0374	100.0	10300	0.0111	0.7860	0.9662	0.9953	nan	0.9972	0.9914	0.8667	0.9770	0.9987	0.9962	0.9827	0.7825	0.9593	0.9954

Framework versions

Transformers 4.33.1
Pytorch 2.0.1+cu118
Datasets 2.14.5
Tokenizers 0.13.3

segformer-b0_DsB1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NSDT 3DConvert

UnrealSynth

DreamTexture.js