earthquake

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

• Loss: 1.2868
• Mean Iou: 0.2379
• Mean Accuracy: 0.2922
• Overall Accuracy: 0.5492
• Accuracy Background: nan
• Accuracy Car: 0.0
• Accuracy Earthquake-roads: 0.4334
• Accuracy Other: 0.1291
• Accuracy Road: 0.6209
• Accuracy Road-cracks: 0.0
• Accuracy Sky: 0.8620
• Accuracy Wall: 0.0
• Iou Background: 0.0
• Iou Car: 0.0
• Iou Earthquake-roads: 0.3329
• Iou Other: 0.1229
• Iou Road: 0.5861
• Iou Road-cracks: 0.0
• Iou Sky: 0.8615
• Iou Wall: 0.0

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: 0.0001
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Background	Accuracy Car	Accuracy Earthquake-roads	Accuracy Other	Accuracy Road	Accuracy Road-cracks	Accuracy Sky	Accuracy Wall	Iou Background	Iou Car	Iou Earthquake-roads	Iou Other	Iou Road	Iou Road-cracks	Iou Sky	Iou Wall
2.0651	0.1316	5	2.0768	0.0374	0.1237	0.1091	nan	0.0	0.6207	0.2450	0.0	0.0	0.0	0.0	0.0	0.0	0.1556	0.1438	0.0	0.0	0.0	0.0
1.8605	0.2632	10	2.0532	0.0563	0.1524	0.2410	nan	0.0	0.7513	0.1438	0.1716	0.0	0.0	0.0	0.0	0.0	0.2042	0.0782	0.1683	0.0	0.0	0.0
1.7033	0.3947	15	2.0117	0.0892	0.1338	0.3878	nan	0.0	0.4356	0.0185	0.4531	0.0	0.0296	0.0	0.0	0.0	0.2332	0.0152	0.4359	0.0	0.0296	0.0
1.6843	0.5263	20	1.9639	0.1220	0.2025	0.4637	nan	0.0	0.6874	0.0325	0.4990	0.0	0.1983	0.0	0.0	0.0	0.2913	0.0309	0.4793	0.0	0.1748	0.0
1.7298	0.6579	25	1.8835	0.1234	0.1819	0.5114	nan	0.0	0.6336	0.0555	0.5840	0.0	0.0	0.0	0.0	0.0	0.3818	0.0542	0.5516	0.0	0.0	0.0
1.4833	0.7895	30	1.7512	0.1555	0.2118	0.6238	nan	0.0	0.6148	0.1323	0.7352	0.0	0.0	0.0	0.0	0.0	0.4310	0.1293	0.6841	0.0	0.0	0.0
1.5078	0.9211	35	1.6947	0.2232	0.3260	0.6338	nan	0.0	0.6497	0.4078	0.6786	0.0	0.5459	0.0	0.0	0.0	0.4381	0.3286	0.6399	0.0	0.3788	0.0
1.2623	1.0526	40	1.6373	0.2083	0.3221	0.5803	nan	0.0	0.6784	0.3435	0.6017	0.0	0.6309	0.0	0.0	0.0	0.4245	0.2883	0.5717	0.0	0.3819	0.0
1.124	1.1842	45	1.4797	0.2175	0.3017	0.6133	nan	0.0	0.6023	0.2190	0.6795	0.0	0.6111	0.0	0.0	0.0	0.4482	0.2035	0.6431	0.0	0.4454	0.0
1.1958	1.3158	50	1.3304	0.2445	0.3053	0.6530	nan	0.0	0.5340	0.2660	0.7434	0.0	0.5935	0.0	0.0	0.0	0.4644	0.2522	0.6977	0.0	0.5413	0.0
1.6012	1.4474	55	1.2994	0.2460	0.3082	0.6439	nan	0.0	0.5639	0.3192	0.7202	0.0	0.5538	0.0	0.0	0.0	0.4683	0.3049	0.6765	0.0	0.5182	0.0
0.8773	1.5789	60	1.3010	0.2476	0.3222	0.6103	nan	0.0	0.6028	0.3035	0.6603	0.0	0.6891	0.0	0.0	0.0	0.4475	0.2968	0.6193	0.0	0.6169	0.0
1.2279	1.7105	65	1.4635	0.1929	0.3179	0.4692	nan	0.0	0.6939	0.2394	0.4448	0.0	0.8469	0.0	0.0	0.0	0.3517	0.2280	0.4246	0.0	0.5389	0.0
1.2859	1.8421	70	1.4540	0.2030	0.2891	0.4708	nan	0.0	0.5973	0.1369	0.4808	0.0	0.8089	0.0	0.0	0.0	0.4248	0.1346	0.4546	0.0	0.6098	0.0
1.6452	1.9737	75	1.4244	0.2336	0.3038	0.5933	nan	0.0	0.5942	0.1817	0.6531	0.0	0.6979	0.0	0.0	0.0	0.4350	0.1742	0.6117	0.0	0.6476	0.0
0.9851	2.1053	80	1.2722	0.2418	0.3033	0.6973	nan	0.0	0.4978	0.0674	0.8303	0.0	0.7273	0.0	0.0	0.0	0.4238	0.0668	0.7747	0.0	0.6693	0.0
0.9787	2.2368	85	1.2686	0.2403	0.2910	0.6700	nan	0.0	0.4831	0.0065	0.8018	0.0	0.7458	0.0	0.0	0.0	0.4308	0.0064	0.7577	0.0	0.7276	0.0
0.8809	2.3684	90	1.2555	0.2511	0.3109	0.6567	nan	0.0	0.6051	0.0292	0.7541	0.0	0.7878	0.0	0.0	0.0	0.5059	0.0288	0.7246	0.0	0.7497	0.0
1.0033	2.5	95	1.2597	0.2597	0.3421	0.6828	nan	0.0	0.7784	0.0607	0.7517	0.0	0.8040	0.0	0.0	0.0	0.5432	0.0594	0.7306	0.0	0.7441	0.0
0.9275	2.6316	100	1.2301	0.2695	0.3516	0.7156	nan	0.0	0.7590	0.1137	0.7954	0.0	0.7928	0.0	0.0	0.0	0.5258	0.1102	0.7662	0.0	0.7541	0.0
0.8305	2.7632	105	1.1418	0.2914	0.3563	0.7436	nan	0.0	0.5627	0.2494	0.8540	0.0	0.8279	0.0	0.0	0.0	0.4717	0.2316	0.8070	0.0	0.8207	0.0
1.4194	2.8947	110	1.2171	0.2577	0.3141	0.6741	nan	0.0	0.3442	0.2353	0.8020	0.0	0.8172	0.0	0.0	0.0	0.3208	0.2179	0.7558	0.0	0.7671	0.0
0.9159	3.0263	115	1.5539	0.1992	0.2485	0.3971	nan	0.0	0.3474	0.1528	0.4246	0.0	0.8146	0.0	0.0	0.0	0.2509	0.1430	0.4104	0.0	0.7895	0.0
0.8513	3.1579	120	1.6732	0.1939	0.2713	0.3645	nan	0.0	0.6345	0.1103	0.3273	0.0	0.8272	0.0	0.0	0.0	0.3363	0.1081	0.3223	0.0	0.7842	0.0
0.655	3.2895	125	1.3317	0.2348	0.3000	0.5930	nan	0.0	0.4572	0.1112	0.6802	0.0	0.8512	0.0	0.0	0.0	0.3833	0.1091	0.6360	0.0	0.7501	0.0
1.1348	3.4211	130	1.2073	0.2544	0.3232	0.6840	nan	0.0	0.5067	0.0727	0.8010	0.0	0.8817	0.0	0.0	0.0	0.4064	0.0715	0.7402	0.0	0.8169	0.0
1.0625	3.5526	135	1.2138	0.2499	0.3055	0.6480	nan	0.0	0.4694	0.0793	0.7599	0.0	0.8299	0.0	0.0	0.0	0.3803	0.0778	0.7140	0.0	0.8275	0.0
0.9589	3.6842	140	1.3193	0.2404	0.3097	0.6012	nan	0.0	0.6269	0.0484	0.6678	0.0	0.8249	0.0	0.0	0.0	0.4164	0.0478	0.6404	0.0	0.8189	0.0
1.246	3.8158	145	1.3526	0.2097	0.2537	0.5136	nan	0.0	0.3748	0.0373	0.5993	0.0	0.7645	0.0	0.0	0.0	0.3062	0.0366	0.5704	0.0	0.7645	0.0
0.7836	3.9474	150	1.3824	0.1970	0.2338	0.4868	nan	0.0	0.2695	0.0250	0.5844	0.0	0.7579	0.0	0.0	0.0	0.2407	0.0248	0.5525	0.0	0.7579	0.0
0.8474	4.0789	155	1.4235	0.1998	0.2368	0.4725	nan	0.0	0.2587	0.0655	0.5606	0.0	0.7724	0.0	0.0	0.0	0.2365	0.0629	0.5275	0.0	0.7712	0.0
1.0723	4.2105	160	1.4177	0.2109	0.2510	0.4877	nan	0.0	0.2976	0.1024	0.5688	0.0	0.7884	0.0	0.0	0.0	0.2681	0.0978	0.5343	0.0	0.7868	0.0
1.1283	4.3421	165	1.4844	0.2020	0.2396	0.4298	nan	0.0	0.2979	0.1040	0.4875	0.0	0.7877	0.0	0.0	0.0	0.2671	0.1007	0.4621	0.0	0.7860	0.0
0.6614	4.4737	170	1.4177	0.2111	0.2522	0.4542	nan	0.0	0.3361	0.1011	0.5131	0.0	0.8151	0.0	0.0	0.0	0.2906	0.0980	0.4868	0.0	0.8136	0.0
1.0973	4.6053	175	1.5041	0.2078	0.2573	0.4288	nan	0.0	0.4572	0.0763	0.4572	0.0	0.8107	0.0	0.0	0.0	0.3406	0.0750	0.4386	0.0	0.8081	0.0
0.8756	4.7368	180	1.3542	0.2236	0.2753	0.4929	nan	0.0	0.4403	0.1089	0.5448	0.0	0.8332	0.0	0.0	0.0	0.3338	0.1051	0.5177	0.0	0.8324	0.0
0.6712	4.8684	185	1.3772	0.2232	0.2761	0.4811	nan	0.0	0.4525	0.1373	0.5231	0.0	0.8195	0.0	0.0	0.0	0.3373	0.1303	0.4993	0.0	0.8185	0.0
1.2096	5.0	190	1.2868	0.2379	0.2922	0.5492	nan	0.0	0.4334	0.1291	0.6209	0.0	0.8620	0.0	0.0	0.0	0.3329	0.1229	0.5861	0.0	0.8615	0.0

Framework versions

• Transformers 4.46.2
• Pytorch 2.4.1+cpu
• Datasets 3.1.0
• Tokenizers 0.20.3