segformer-b0-finetuned-segments-sidewalk-2

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

Loss: 0.8968
Mean Iou: 0.1704
Mean Accuracy: 0.2137
Overall Accuracy: 0.7831
Accuracy Unlabeled: nan
Accuracy Flat-road: 0.8436
Accuracy Flat-sidewalk: 0.9541
Accuracy Flat-crosswalk: 0.0
Accuracy Flat-cyclinglane: 0.5543
Accuracy Flat-parkingdriveway: 0.0590
Accuracy Flat-railtrack: nan
Accuracy Flat-curb: 0.0004
Accuracy Human-person: 0.0
Accuracy Human-rider: 0.0
Accuracy Vehicle-car: 0.8749
Accuracy Vehicle-truck: 0.0
Accuracy Vehicle-bus: 0.0
Accuracy Vehicle-tramtrain: 0.0
Accuracy Vehicle-motorcycle: 0.0
Accuracy Vehicle-bicycle: 0.0
Accuracy Vehicle-caravan: 0.0
Accuracy Vehicle-cartrailer: 0.0
Accuracy Construction-building: 0.8927
Accuracy Construction-door: 0.0
Accuracy Construction-wall: 0.0002
Accuracy Construction-fenceguardrail: 0.0
Accuracy Construction-bridge: 0.0
Accuracy Construction-tunnel: nan
Accuracy Construction-stairs: 0.0
Accuracy Object-pole: 0.0
Accuracy Object-trafficsign: 0.0
Accuracy Object-trafficlight: 0.0
Accuracy Nature-vegetation: 0.9196
Accuracy Nature-terrain: 0.8296
Accuracy Sky: 0.9107
Accuracy Void-ground: 0.0
Accuracy Void-dynamic: 0.0
Accuracy Void-static: 0.0
Accuracy Void-unclear: 0.0
Iou Unlabeled: nan
Iou Flat-road: 0.5937
Iou Flat-sidewalk: 0.8167
Iou Flat-crosswalk: 0.0
Iou Flat-cyclinglane: 0.4872
Iou Flat-parkingdriveway: 0.0549
Iou Flat-railtrack: nan
Iou Flat-curb: 0.0004
Iou Human-person: 0.0
Iou Human-rider: 0.0
Iou Vehicle-car: 0.6327
Iou Vehicle-truck: 0.0
Iou Vehicle-bus: 0.0
Iou Vehicle-tramtrain: 0.0
Iou Vehicle-motorcycle: 0.0
Iou Vehicle-bicycle: 0.0
Iou Vehicle-caravan: 0.0
Iou Vehicle-cartrailer: 0.0
Iou Construction-building: 0.6091
Iou Construction-door: 0.0
Iou Construction-wall: 0.0002
Iou Construction-fenceguardrail: 0.0
Iou Construction-bridge: 0.0
Iou Construction-tunnel: nan
Iou Construction-stairs: 0.0
Iou Object-pole: 0.0
Iou Object-trafficsign: 0.0
Iou Object-trafficlight: 0.0
Iou Nature-vegetation: 0.7798
Iou Nature-terrain: 0.6372
Iou Sky: 0.8419
Iou Void-ground: 0.0
Iou Void-dynamic: 0.0
Iou Void-static: 0.0
Iou Void-unclear: 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: 6e-05
train_batch_size: 2
eval_batch_size: 2
seed: 42
optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Flat-road	Accuracy Flat-sidewalk	Accuracy Flat-crosswalk	Accuracy Flat-cyclinglane	Accuracy Flat-parkingdriveway	Accuracy Flat-railtrack	Accuracy Flat-curb	Accuracy Vehicle-car	Accuracy Vehicle-bicycle	Accuracy Construction-building	Accuracy Construction-wall	Accuracy Construction-tunnel	Accuracy Object-pole	Accuracy Nature-vegetation	Accuracy Nature-terrain	Accuracy Sky	Accuracy Void-static	Iou Unlabeled	Iou Flat-road	Iou Flat-sidewalk	Iou Flat-crosswalk	Iou Flat-cyclinglane	Iou Flat-parkingdriveway	Iou Flat-railtrack	Iou Flat-curb	Iou Vehicle-car	Iou Vehicle-bicycle	Iou Construction-building	Iou Construction-wall	Iou Construction-tunnel	Iou Object-pole	Iou Nature-vegetation	Iou Nature-terrain	Iou Sky	Iou Void-static
2.9768	0.05	20	3.1791	0.0824	0.1279	0.5913	nan	0.0157	0.9490	0.0000	0.0746	0.0066	nan	0.0002	0.7922	0.0101	0.7108	0.0376	nan	0.0005	0.8123	0.0001	0.6566	0.0258	0.0	0.0156	0.6477	0.0000	0.0568	0.0062	0.0	0.0002	0.5028	0.0088	0.3687	0.0316	0.0	0.0003	0.6234	0.0001	0.6052	0.0169
2.6052	0.1	40	2.4058	0.0887	0.1358	0.6234	nan	0.0760	0.9605	0.0	0.1023	0.0053	nan	0.0000	0.9062	0.0	0.6993	0.0163	nan	0.0016	0.9314	0.0155	0.6300	0.0013	nan	0.0735	0.6428	0.0	0.0875	0.0051	0.0	0.0000	0.3827	0.0	0.4673	0.0159	nan	0.0015	0.6230	0.0154	0.6121	0.0013
2.1705	0.15	60	2.0976	0.1035	0.1501	0.6688	nan	0.4365	0.9532	0.0000	0.0856	0.0037	nan	0.0000	0.8058	0.0	0.8240	0.0166	nan	0.0	0.9208	0.0568	0.7008	0.0003	nan	0.3480	0.7087	0.0000	0.0843	0.0036	0.0	0.0000	0.4109	0.0	0.4710	0.0163	nan	0.0	0.6526	0.0546	0.6640	0.0003
1.8745	0.2	80	1.8802	0.1110	0.1593	0.6765	nan	0.4471	0.9504	0.0	0.0767	0.0010	nan	0.0	0.9217	0.0	0.7561	0.0224	nan	0.0	0.9143	0.2378	0.7690	0.0	nan	0.3356	0.7166	0.0	0.0760	0.0010	0.0	0.0	0.3837	0.0	0.5042	0.0220	nan	0.0	0.6802	0.2066	0.7370	0.0
2.1542	0.25	100	1.7926	0.1125	0.1595	0.6886	nan	0.6494	0.9213	0.0	0.0678	0.0040	nan	0.0	0.7104	0.0	0.8395	0.0025	nan	0.0	0.9571	0.1054	0.8468	0.0	nan	0.4172	0.7575	0.0	0.0667	0.0039	0.0	0.0	0.5175	0.0	0.4994	0.0025	nan	0.0	0.6264	0.0870	0.7340	0.0
1.6841	0.3	120	1.5504	0.1192	0.1621	0.6997	nan	0.6287	0.9471	0.0	0.0446	0.0035	nan	0.0	0.8139	0.0	0.8574	0.0017	nan	0.0	0.9511	0.0879	0.8510	0.0	nan	0.4446	0.7485	0.0	0.0445	0.0034	nan	0.0	0.5555	0.0	0.5313	0.0017	nan	0.0	0.6343	0.0796	0.7711	0.0
2.1305	0.35	140	1.5125	0.1353	0.1784	0.7237	nan	0.6944	0.9571	0.0	0.0037	0.0017	nan	0.0	0.8413	0.0	0.8738	0.0025	nan	0.0	0.8910	0.5530	0.8888	0.0	nan	0.4537	0.7533	0.0	0.0037	0.0017	nan	0.0	0.5602	0.0	0.5346	0.0025	nan	0.0	0.7488	0.4770	0.7944	0.0
1.6506	0.4	160	1.3938	0.1218	0.1654	0.7090	nan	0.7008	0.9472	0.0	0.0038	0.0014	nan	0.0	0.8214	0.0	0.8928	0.0001	nan	0.0	0.9339	0.1333	0.8579	0.0	nan	0.4598	0.7695	0.0	0.0038	0.0014	nan	0.0	0.5550	0.0	0.5385	0.0001	nan	0.0	0.6621	0.1210	0.7865	0.0
1.7185	0.45	180	1.3990	0.1329	0.1735	0.7166	nan	0.5564	0.9762	0.0	0.0119	0.0012	nan	0.0000	0.8480	0.0	0.8853	0.0	nan	0.0	0.9079	0.4818	0.8823	0.0	nan	0.4388	0.7150	0.0	0.0119	0.0012	nan	0.0000	0.5632	0.0	0.5472	0.0	nan	0.0	0.7402	0.4295	0.8062	0.0
1.7036	0.5	200	1.3011	0.1287	0.1736	0.7164	nan	0.7953	0.9197	0.0	0.0628	0.0021	nan	0.0	0.8276	0.0	0.8438	0.0	nan	0.0	0.9521	0.2419	0.9091	0.0	nan	0.4573	0.7800	0.0	0.0627	0.0021	nan	0.0	0.5829	0.0	0.5572	0.0	nan	0.0	0.6722	0.2130	0.7900	0.0
1.5361	0.55	220	1.2613	0.1432	0.1874	0.7365	nan	0.6906	0.9669	0.0	0.1806	0.0011	nan	0.0	0.9072	0.0	0.8617	0.0	nan	0.0	0.8740	0.6097	0.9044	0.0	nan	0.4911	0.7595	0.0	0.1798	0.0011	nan	0.0	0.5114	0.0	0.5596	0.0	nan	0.0	0.7569	0.5106	0.8120	0.0
1.6957	0.6	240	1.2449	0.1518	0.1952	0.7519	nan	0.7923	0.9434	0.0	0.4161	0.0018	nan	0.0	0.8229	0.0	0.8715	0.0	nan	0.0	0.9295	0.5466	0.9232	0.0	nan	0.5281	0.7932	0.0	0.3983	0.0018	nan	0.0	0.5884	0.0	0.5666	0.0	nan	0.0	0.7195	0.4723	0.7883	0.0
1.2337	0.65	260	1.1976	0.1469	0.1911	0.7477	nan	0.8007	0.9428	0.0	0.4690	0.0049	nan	0.0	0.8529	0.0	0.8846	0.0	nan	0.0	0.9327	0.3037	0.9238	0.0	nan	0.5481	0.7980	0.0	0.4328	0.0049	nan	0.0	0.5794	0.0	0.5667	0.0	nan	0.0	0.6942	0.2785	0.7994	0.0
1.4384	0.7	280	1.1669	0.1586	0.2020	0.7582	nan	0.7685	0.9574	0.0	0.4655	0.0083	nan	0.0	0.8418	0.0	0.9012	0.0	nan	0.0	0.8563	0.7800	0.8842	0.0	nan	0.5481	0.7808	0.0	0.4389	0.0083	nan	0.0	0.6082	0.0	0.5640	0.0	nan	0.0	0.7510	0.5589	0.8169	0.0
1.4494	0.75	300	1.1346	0.1576	0.2019	0.7591	nan	0.7665	0.9456	0.0	0.4582	0.0060	nan	0.0	0.8569	0.0	0.8554	0.0	nan	0.0	0.9294	0.7121	0.9308	0.0	nan	0.5282	0.7845	0.0	0.4217	0.0059	nan	0.0	0.5847	0.0	0.5850	0.0	nan	0.0	0.7492	0.5904	0.7944	0.0
0.8967	0.8	320	1.1276	0.1605	0.2064	0.7652	nan	0.8038	0.9449	0.0	0.4669	0.0098	nan	0.0	0.8488	0.0	0.8895	0.0	nan	0.0	0.8921	0.8178	0.9316	0.0	nan	0.5427	0.7987	0.0	0.4375	0.0097	nan	0.0	0.5922	0.0	0.5795	0.0	nan	0.0	0.7703	0.6055	0.7996	0.0
1.4394	0.85	340	1.0325	0.1621	0.2066	0.7685	nan	0.8425	0.9359	0.0	0.4954	0.0061	nan	0.0	0.8497	0.0	0.8871	0.0	nan	0.0	0.9215	0.7818	0.8913	0.0	nan	0.5464	0.8116	0.0	0.4548	0.0061	nan	0.0	0.6081	0.0	0.5849	0.0	nan	0.0	0.7582	0.6066	0.8093	0.0
1.6443	0.9	360	1.0352	0.1627	0.2064	0.7689	nan	0.8406	0.9391	0.0	0.4941	0.0133	nan	0.0	0.8390	0.0	0.8895	0.0	nan	0.0	0.9223	0.7547	0.9115	0.0	nan	0.5590	0.8059	0.0	0.4460	0.0131	nan	0.0	0.6100	0.0	0.5802	0.0	nan	0.0	0.7595	0.6157	0.8155	0.0
0.8427	0.95	380	1.0465	0.1617	0.2052	0.7673	nan	0.7660	0.9577	0.0	0.4633	0.0144	nan	0.0	0.8492	0.0	0.9211	0.0	nan	0.0	0.8976	0.7676	0.9286	0.0	nan	0.5603	0.7982	0.0	0.4237	0.0142	nan	0.0	0.6032	0.0	0.5604	0.0	nan	0.0	0.7779	0.6173	0.8188	0.0
1.0543	1.0	400	1.0080	0.1634	0.2095	0.7708	nan	0.8675	0.9315	0.0	0.5030	0.0110	nan	0.0	0.8841	0.0	0.8912	0.0	nan	0.0	0.9024	0.8038	0.9107	0.0	nan	0.5440	0.8189	0.0	0.4477	0.0108	nan	0.0	0.5834	0.0	0.5881	0.0	nan	0.0	0.7807	0.6352	0.8201	0.0
1.4482	1.05	420	0.9947	0.1643	0.2078	0.7733	nan	0.8037	0.9544	0.0	0.5089	0.0197	nan	0.0	0.8718	0.0	0.8994	0.0	nan	0.0	0.9180	0.7525	0.9204	0.0	nan	0.5794	0.8020	0.0	0.4431	0.0193	nan	0.0	0.6047	0.0	0.5886	0.0	nan	0.0	0.7742	0.6271	0.8204	0.0
1.1712	1.1	440	1.0049	0.1650	0.2093	0.7750	nan	0.8328	0.9552	0.0	0.4946	0.0187	nan	0.0	0.8994	0.0	0.8947	0.0	nan	0.0	0.8947	0.7887	0.9185	0.0	nan	0.5809	0.8057	0.0	0.4409	0.0182	nan	0.0	0.5872	0.0	0.6006	0.0	nan	0.0	0.7813	0.6348	0.8301	0.0
0.9807	1.15	460	0.9873	0.1635	0.2064	0.7687	nan	0.7494	0.9591	0.0	0.5054	0.0204	nan	0.0	0.8399	0.0	0.9211	0.0	nan	0.0	0.9021	0.7923	0.9136	0.0	nan	0.5895	0.7875	0.0	0.4247	0.0199	nan	0.0	0.6169	0.0	0.5633	0.0	nan	0.0	0.7763	0.6261	0.8286	0.0
1.2975	1.2	480	0.9780	0.1642	0.2094	0.7727	nan	0.8164	0.9528	0.0	0.4900	0.0279	nan	0.0	0.8914	0.0	0.8669	0.0	nan	0.0	0.9054	0.8257	0.9226	0.0	nan	0.5758	0.7996	0.0	0.4333	0.0269	nan	0.0	0.5915	0.0	0.6088	0.0	nan	0.0	0.7704	0.6187	0.8281	0.0
0.8231	1.25	500	0.9648	0.1651	0.2110	0.7740	nan	0.8022	0.9458	0.0	0.5455	0.0325	nan	0.0000	0.8859	0.0	0.9026	0.0	nan	0.0	0.9088	0.8250	0.9044	0.0	nan	0.5830	0.8105	0.0	0.4259	0.0315	nan	0.0000	0.5989	0.0	0.5913	0.0	nan	0.0	0.7738	0.6307	0.8378	0.0
0.7362	1.3	520	0.9449	0.1670	0.2123	0.7767	nan	0.8147	0.9465	0.0	0.5651	0.0467	nan	0.0000	0.8541	0.0	0.8977	0.0	nan	0.0	0.9154	0.8255	0.9290	0.0	nan	0.5862	0.8114	0.0	0.4447	0.0446	nan	0.0000	0.6331	0.0	0.5934	0.0	nan	0.0	0.7739	0.6216	0.8349	0.0
1.0454	1.35	540	0.9437	0.1680	0.2114	0.7778	nan	0.8559	0.9498	0.0	0.5092	0.0576	nan	0.0	0.8576	0.0	0.9108	0.0	nan	0.0	0.8943	0.8168	0.9134	0.0	nan	0.5826	0.8136	0.0	0.4673	0.0536	nan	0.0	0.6262	0.0	0.5853	0.0	nan	0.0	0.7848	0.6312	0.8314	0.0
1.0599	1.4	560	0.9566	0.1677	0.2112	0.7794	nan	0.8449	0.9563	0.0	0.5049	0.0493	nan	0.0000	0.8986	0.0	0.8723	0.0	nan	0.0	0.9172	0.7992	0.9157	0.0	nan	0.5873	0.8128	0.0	0.4601	0.0460	nan	0.0000	0.5953	0.0	0.6089	0.0	nan	0.0	0.7800	0.6341	0.8428	0.0
0.6598	1.45	580	0.9155	0.1688	0.2111	0.7791	nan	0.8050	0.9622	0.0	0.5365	0.0473	nan	0.0001	0.8509	0.0	0.9171	0.0	nan	0.0	0.9023	0.8105	0.9243	0.0	nan	0.6049	0.8057	0.0	0.4731	0.0445	nan	0.0001	0.6307	0.0	0.5770	0.0	nan	0.0	0.7859	0.6391	0.8399	0.0
0.7654	1.5	600	0.9185	0.1687	0.2130	0.7796	nan	0.8068	0.9550	0.0	0.5702	0.0536	nan	0.0001	0.8729	0.0	0.8931	0.0	nan	0.0	0.9151	0.8346	0.9148	0.0	nan	0.5971	0.8098	0.0	0.4719	0.0500	nan	0.0001	0.6133	0.0	0.5988	0.0	nan	0.0	0.7789	0.6372	0.8410	0.0
0.8716	1.55	620	0.9162	0.1682	0.2139	0.7785	nan	0.8133	0.9584	0.0	0.5727	0.0477	nan	0.0000	0.8838	0.0	0.9012	0.0	nan	0.0	0.8745	0.8665	0.9252	0.0	nan	0.6013	0.8067	0.0	0.4779	0.0446	nan	0.0000	0.6182	0.0	0.6048	0.0	nan	0.0	0.7763	0.6088	0.8440	0.0
1.318	1.6	640	0.9260	0.1689	0.2120	0.7789	nan	0.7813	0.9632	0.0	0.5634	0.0477	nan	0.0000	0.8681	0.0	0.8929	0.0	nan	0.0	0.9127	0.8219	0.9338	0.0	nan	0.6042	0.7966	0.0	0.4703	0.0448	nan	0.0000	0.6301	0.0	0.6015	0.0	nan	0.0	0.7769	0.6408	0.8411	0.0
0.9591	1.65	660	0.9280	0.1698	0.2116	0.7801	nan	0.7964	0.9618	0.0	0.5694	0.0469	nan	0.0017	0.8496	0.0	0.9150	0.0	nan	0.0	0.9129	0.7988	0.9186	0.0	nan	0.6111	0.8011	0.0	0.4809	0.0443	nan	0.0017	0.6406	0.0	0.5848	0.0	nan	0.0	0.7832	0.6439	0.8435	0.0
2.0507	1.7	680	0.8976	0.1706	0.2147	0.7827	nan	0.8522	0.9455	0.0	0.5886	0.0577	nan	0.0004	0.8642	0.0	0.9091	0.0000	nan	0.0	0.9176	0.8095	0.9253	0.0	nan	0.5923	0.8223	0.0	0.4984	0.0538	nan	0.0004	0.6331	0.0	0.5927	0.0000	nan	0.0	0.7836	0.6421	0.8416	0.0
1.3324	1.75	700	0.9057	0.1700	0.2138	0.7812	nan	0.8051	0.9585	0.0	0.5747	0.0571	nan	0.0010	0.8788	0.0	0.9131	0.0	nan	0.0	0.9019	0.8239	0.9280	0.0	nan	0.6084	0.8069	0.0	0.4912	0.0533	nan	0.0010	0.6241	0.0	0.5921	0.0	nan	0.0	0.7868	0.6351	0.8416	0.0
1.4375	1.8	720	0.8915	0.1700	0.2139	0.7821	nan	0.8257	0.9584	0.0	0.5754	0.0480	nan	0.0003	0.8833	0.0	0.8923	0.0002	nan	0.0	0.9061	0.8238	0.9310	0.0	nan	0.6000	0.8091	0.0	0.4905	0.0452	nan	0.0003	0.6249	0.0	0.6081	0.0002	nan	0.0	0.7838	0.6374	0.8401	0.0
1.3081	1.85	740	0.9020	0.1705	0.2151	0.7830	nan	0.8329	0.9552	0.0	0.5666	0.0659	nan	0.0022	0.8832	0.0	0.8979	0.0001	nan	0.0	0.9047	0.8435	0.9299	0.0	nan	0.6022	0.8146	0.0	0.4910	0.0608	nan	0.0022	0.6248	0.0	0.6061	0.0001	nan	0.0	0.7829	0.6327	0.8402	0.0
0.9699	1.9	760	0.8899	0.1707	0.2143	0.7823	nan	0.8023	0.9603	0.0	0.5820	0.0633	nan	0.0067	0.8854	0.0	0.9014	0.0	nan	0.0	0.9102	0.8198	0.9266	0.0	nan	0.6117	0.8065	0.0	0.4926	0.0588	nan	0.0066	0.6189	0.0	0.6022	0.0	nan	0.0	0.7832	0.6394	0.8423	0.0
1.4962	1.95	780	0.8850	0.1702	0.2141	0.7829	nan	0.8367	0.9554	0.0	0.5611	0.0611	nan	0.0009	0.8816	0.0	0.8737	0.0010	nan	0.0	0.9252	0.8238	0.9297	0.0	nan	0.5985	0.8144	0.0	0.4894	0.0568	nan	0.0009	0.6240	0.0	0.6142	0.0010	nan	0.0	0.7734	0.6355	0.8383	0.0
1.0391	2.0	800	0.8968	0.1704	0.2137	0.7831	nan	0.8436	0.9541	0.0	0.5543	0.0590	nan	0.0004	0.8749	0.0	0.8927	0.0002	nan	0.0	0.9196	0.8296	0.9107	0.0	nan	0.5937	0.8167	0.0	0.4872	0.0549	nan	0.0004	0.6327	0.0	0.6091	0.0002	nan	0.0	0.7798	0.6372	0.8419	0.0

Framework versions

Transformers 4.57.3
Pytorch 2.9.1+cu128
Datasets 4.4.1
Tokenizers 0.22.1

Downloads last month: 8

Safetensors

Model size

3.72M params

Tensor type

F32

Model tree for balzj/segformer-b0-finetuned-segments-sidewalk-2

Base model

nvidia/mit-b0

Finetuned

(460)

this model

balzj
/

segformer-b0-finetuned-segments-sidewalk-2

segformer-b0-finetuned-segments-sidewalk-2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Model tree for balzj/segformer-b0-finetuned-segments-sidewalk-2

Evaluation results