parking-terrain

sam1120/parking-terrain

parking-terrain is a image segmentation model from sam1120. Use it for the image segmentation task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as other.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

7% of all-time downloads

All-time downloads

740

Public

Repo size

30.8 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.pt2.7 GB · 62%

At a glance

Task: Image Segmentation
Library: transformers
License: other
Model type: segformer
Access: Public
Created: Feb 25, 2024
Updated: Feb 26, 2024
SHA: f1f1784e

Model card

From the Hugging Face model README

parking-terrain

This model is a fine-tuned version of nvidia/mit-b5 on the sam1120/parking-terrain dataset. It achieves the following results on the evaluation set:

Loss: 0.0226
Mean Iou: 0.9423
Mean Accuracy: 0.9658
Overall Accuracy: 0.9939
Accuracy Unlabeled: nan
Accuracy Sidewalk: 0.9057
Accuracy Road: 0.9946
Accuracy Else: 0.9971
Iou Unlabeled: nan
Iou Sidewalk: 0.8453
Iou Road: 0.9880
Iou Else: 0.9938

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: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.05
num_epochs: 600

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Sidewalk	Accuracy Road	Accuracy Else	Iou Unlabeled	Iou Sidewalk	Iou Road	Iou Else
1.0322	5.0	20	1.1061	0.4272	0.7323	0.7757	nan	0.5362	0.9371	0.7236	0.0	0.4020	0.6015	0.7052
0.3821	10.0	40	0.3327	0.8052	0.8879	0.9543	nan	0.7204	0.9958	0.9475	nan	0.5966	0.8762	0.9429
0.1146	15.0	60	0.0906	0.8490	0.8945	0.9790	nan	0.7050	0.9951	0.9833	nan	0.6242	0.9436	0.9793
0.0668	20.0	80	0.0558	0.8717	0.8905	0.9852	nan	0.6823	0.9966	0.9925	nan	0.6637	0.9670	0.9844
0.0491	25.0	100	0.0474	0.8892	0.9372	0.9863	nan	0.8296	0.9916	0.9903	nan	0.7112	0.9708	0.9855
0.04	30.0	120	0.0366	0.9238	0.9542	0.9899	nan	0.8776	0.9914	0.9937	nan	0.8068	0.9755	0.9890
0.0339	35.0	140	0.0332	0.9263	0.9656	0.9906	nan	0.9130	0.9901	0.9937	nan	0.8096	0.9801	0.9891
0.0341	40.0	160	0.0310	0.9266	0.9653	0.9905	nan	0.9130	0.9889	0.9941	nan	0.8118	0.9782	0.9897
0.0295	45.0	180	0.0253	0.9407	0.9675	0.9923	nan	0.9168	0.9893	0.9963	nan	0.8486	0.9822	0.9913
0.058	50.0	200	0.0262	0.9312	0.9645	0.9918	nan	0.9076	0.9902	0.9956	nan	0.8211	0.9809	0.9916
0.0242	55.0	220	0.0248	0.9377	0.9645	0.9925	nan	0.9051	0.9923	0.9960	nan	0.8376	0.9838	0.9918
0.0217	60.0	240	0.0250	0.9324	0.9662	0.9920	nan	0.9128	0.9897	0.9960	nan	0.8229	0.9829	0.9915
0.0223	65.0	260	0.0248	0.9339	0.9598	0.9919	nan	0.8920	0.9916	0.9959	nan	0.8281	0.9826	0.9909
0.0206	70.0	280	0.0215	0.9411	0.9643	0.9929	nan	0.9038	0.9926	0.9965	nan	0.8466	0.9847	0.9921
0.0194	75.0	300	0.0226	0.9351	0.9669	0.9924	nan	0.9126	0.9927	0.9954	nan	0.8289	0.9849	0.9915
0.02	80.0	320	0.0216	0.9381	0.9626	0.9927	nan	0.8993	0.9918	0.9967	nan	0.8371	0.9854	0.9917
0.0181	85.0	340	0.0222	0.9333	0.9653	0.9923	nan	0.9092	0.9907	0.9961	nan	0.8239	0.9846	0.9915
0.018	90.0	360	0.0225	0.9340	0.9624	0.9924	nan	0.8993	0.9914	0.9964	nan	0.8256	0.9848	0.9916
0.0169	95.0	380	0.0207	0.9361	0.9663	0.9928	nan	0.9099	0.9931	0.9958	nan	0.8306	0.9855	0.9922
0.0157	100.0	400	0.0209	0.9337	0.9658	0.9926	nan	0.9092	0.9923	0.9960	nan	0.8234	0.9858	0.9921
0.0147	105.0	420	0.0229	0.9303	0.9663	0.9921	nan	0.9120	0.9914	0.9954	nan	0.8153	0.9836	0.9918
0.014	110.0	440	0.0205	0.9370	0.9658	0.9930	nan	0.9082	0.9930	0.9962	nan	0.8328	0.9857	0.9926
0.013	115.0	460	0.0213	0.9375	0.9613	0.9929	nan	0.8944	0.9928	0.9968	nan	0.8345	0.9857	0.9922
0.0146	120.0	480	0.0201	0.9409	0.9605	0.9935	nan	0.8914	0.9922	0.9979	nan	0.8432	0.9866	0.9930
0.0141	125.0	500	0.0206	0.9359	0.9669	0.9930	nan	0.9110	0.9937	0.9959	nan	0.8285	0.9864	0.9926
0.0146	130.0	520	0.0206	0.9378	0.9629	0.9931	nan	0.8983	0.9942	0.9963	nan	0.8346	0.9863	0.9925
0.0132	135.0	540	0.0206	0.9354	0.9614	0.9931	nan	0.8931	0.9947	0.9963	nan	0.8268	0.9868	0.9926
0.0116	140.0	560	0.0191	0.9387	0.9655	0.9934	nan	0.9062	0.9937	0.9966	nan	0.8358	0.9870	0.9931
0.012	145.0	580	0.0206	0.9384	0.9618	0.9933	nan	0.8954	0.9927	0.9973	nan	0.8357	0.9863	0.9930
0.0131	150.0	600	0.0206	0.9352	0.9666	0.9930	nan	0.9103	0.9937	0.9959	nan	0.8264	0.9866	0.9926
0.0126	155.0	620	0.0205	0.9371	0.9623	0.9931	nan	0.8968	0.9936	0.9966	nan	0.8327	0.9860	0.9927
0.011	160.0	640	0.0197	0.9389	0.9624	0.9935	nan	0.8958	0.9945	0.9968	nan	0.8367	0.9868	0.9933
0.0124	165.0	660	0.0198	0.9378	0.9629	0.9934	nan	0.8982	0.9937	0.9970	nan	0.8331	0.9872	0.9931
0.0119	170.0	680	0.0206	0.9362	0.9648	0.9932	nan	0.9037	0.9943	0.9962	nan	0.8284	0.9872	0.9929
0.0121	175.0	700	0.0210	0.9360	0.9653	0.9931	nan	0.9064	0.9930	0.9965	nan	0.8285	0.9866	0.9929
0.0123	180.0	720	0.0213	0.9340	0.9657	0.9930	nan	0.9069	0.9943	0.9958	nan	0.8229	0.9861	0.9929
0.0098	185.0	740	0.0205	0.9386	0.9644	0.9935	nan	0.9026	0.9936	0.9969	nan	0.8353	0.9872	0.9933
0.0112	190.0	760	0.0204	0.9385	0.9628	0.9935	nan	0.8974	0.9943	0.9969	nan	0.8350	0.9874	0.9932
0.0104	195.0	780	0.0206	0.9368	0.9642	0.9932	nan	0.9021	0.9942	0.9963	nan	0.8310	0.9865	0.9930
0.0112	200.0	800	0.0202	0.9397	0.9647	0.9934	nan	0.9044	0.9926	0.9971	nan	0.8394	0.9867	0.9931
0.0109	205.0	820	0.0201	0.9389	0.9668	0.9935	nan	0.9099	0.9937	0.9967	nan	0.8361	0.9874	0.9933
0.0104	210.0	840	0.0204	0.9394	0.9648	0.9934	nan	0.9041	0.9936	0.9968	nan	0.8383	0.9867	0.9932
0.011	215.0	860	0.0204	0.9392	0.9640	0.9935	nan	0.9006	0.9946	0.9967	nan	0.8371	0.9871	0.9934
0.0103	220.0	880	0.0210	0.9388	0.9642	0.9933	nan	0.9027	0.9929	0.9970	nan	0.8366	0.9867	0.9930
0.0091	225.0	900	0.0220	0.9364	0.9646	0.9931	nan	0.9044	0.9927	0.9967	nan	0.8300	0.9865	0.9927
0.0091	230.0	920	0.0194	0.9423	0.9656	0.9938	nan	0.9058	0.9940	0.9971	nan	0.8457	0.9874	0.9937
0.0089	235.0	940	0.0212	0.9378	0.9646	0.9933	nan	0.9034	0.9937	0.9966	nan	0.8337	0.9868	0.9930
0.008	240.0	960	0.0201	0.9406	0.9665	0.9936	nan	0.9090	0.9933	0.9970	nan	0.8413	0.9869	0.9936
0.0094	245.0	980	0.0211	0.9374	0.9671	0.9932	nan	0.9124	0.9923	0.9966	nan	0.8325	0.9867	0.9929
0.0106	250.0	1000	0.0208	0.9387	0.9669	0.9933	nan	0.9117	0.9921	0.9969	nan	0.8366	0.9864	0.9931
0.0088	255.0	1020	0.0220	0.9354	0.9653	0.9929	nan	0.9062	0.9936	0.9960	nan	0.8271	0.9865	0.9925
0.0091	260.0	1040	0.0205	0.9419	0.9663	0.9937	nan	0.9089	0.9929	0.9972	nan	0.8453	0.9871	0.9934
0.0098	265.0	1060	0.0206	0.9403	0.9647	0.9935	nan	0.9033	0.9943	0.9967	nan	0.8408	0.9870	0.9932
0.0105	270.0	1080	0.0203	0.9398	0.9662	0.9935	nan	0.9083	0.9934	0.9968	nan	0.8392	0.9868	0.9933
0.0088	275.0	1100	0.0209	0.9392	0.9655	0.9935	nan	0.9057	0.9941	0.9966	nan	0.8373	0.9872	0.9932
0.0089	280.0	1120	0.0216	0.9407	0.9659	0.9935	nan	0.9078	0.9928	0.9971	nan	0.8421	0.9867	0.9932
0.0088	285.0	1140	0.0204	0.9390	0.9682	0.9934	nan	0.9150	0.9933	0.9964	nan	0.8370	0.9868	0.9931
0.0088	290.0	1160	0.0215	0.9373	0.9642	0.9934	nan	0.9013	0.9951	0.9963	nan	0.8311	0.9876	0.9931
0.0086	295.0	1180	0.0216	0.9414	0.9648	0.9936	nan	0.9040	0.9933	0.9972	nan	0.8438	0.9869	0.9933
0.0081	300.0	1200	0.0207	0.9404	0.9669	0.9935	nan	0.9106	0.9934	0.9968	nan	0.8410	0.9871	0.9933
0.009	305.0	1220	0.0208	0.9407	0.9660	0.9936	nan	0.9072	0.9939	0.9968	nan	0.8418	0.9872	0.9933
0.0083	310.0	1240	0.0210	0.9416	0.9658	0.9936	nan	0.9069	0.9933	0.9971	nan	0.8444	0.9870	0.9934
0.0094	315.0	1260	0.0206	0.9417	0.9649	0.9938	nan	0.9031	0.9946	0.9969	nan	0.8441	0.9873	0.9936
0.0084	320.0	1280	0.0205	0.9428	0.9645	0.9939	nan	0.9017	0.9946	0.9972	nan	0.8471	0.9875	0.9938
0.0085	325.0	1300	0.0218	0.9384	0.9666	0.9934	nan	0.9093	0.9940	0.9964	nan	0.8348	0.9872	0.9931
0.0089	330.0	1320	0.0207	0.9419	0.9663	0.9937	nan	0.9079	0.9943	0.9968	nan	0.8444	0.9878	0.9934
0.008	335.0	1340	0.0212	0.9420	0.9653	0.9938	nan	0.9045	0.9942	0.9971	nan	0.8446	0.9877	0.9936
0.0076	340.0	1360	0.0211	0.9428	0.9652	0.9939	nan	0.9041	0.9945	0.9972	nan	0.8470	0.9877	0.9938
0.0088	345.0	1380	0.0210	0.9428	0.9654	0.9939	nan	0.9044	0.9948	0.9971	nan	0.8468	0.9880	0.9937
0.0084	350.0	1400	0.0213	0.9412	0.9660	0.9937	nan	0.9068	0.9943	0.9968	nan	0.8425	0.9877	0.9934
0.0077	355.0	1420	0.0215	0.9400	0.9645	0.9937	nan	0.9017	0.9952	0.9966	nan	0.8389	0.9877	0.9935
0.0069	360.0	1440	0.0210	0.9414	0.9664	0.9937	nan	0.9086	0.9936	0.9971	nan	0.8432	0.9874	0.9936
0.0071	365.0	1460	0.0204	0.9436	0.9663	0.9939	nan	0.9083	0.9930	0.9975	nan	0.8496	0.9874	0.9937
0.0079	370.0	1480	0.0208	0.9425	0.9660	0.9939	nan	0.9072	0.9935	0.9974	nan	0.8462	0.9876	0.9937
0.0085	375.0	1500	0.0210	0.9421	0.9661	0.9938	nan	0.9078	0.9934	0.9973	nan	0.8452	0.9875	0.9936
0.0086	380.0	1520	0.0208	0.9425	0.9662	0.9939	nan	0.9078	0.9936	0.9973	nan	0.8462	0.9878	0.9936
0.0076	385.0	1540	0.0218	0.9409	0.9655	0.9936	nan	0.9062	0.9933	0.9971	nan	0.8420	0.9875	0.9933
0.0077	390.0	1560	0.0214	0.9419	0.9656	0.9938	nan	0.9052	0.9945	0.9969	nan	0.8443	0.9879	0.9935
0.0077	395.0	1580	0.0216	0.9405	0.9663	0.9937	nan	0.9075	0.9948	0.9966	nan	0.8400	0.9879	0.9935
0.008	400.0	1600	0.0212	0.9424	0.9658	0.9939	nan	0.9059	0.9945	0.9970	nan	0.8458	0.9879	0.9937
0.0071	405.0	1620	0.0213	0.9412	0.9664	0.9938	nan	0.9078	0.9945	0.9968	nan	0.8423	0.9877	0.9936
0.0082	410.0	1640	0.0218	0.9419	0.9659	0.9938	nan	0.9064	0.9944	0.9970	nan	0.8444	0.9876	0.9936
0.0077	415.0	1660	0.0219	0.9406	0.9664	0.9937	nan	0.9085	0.9937	0.9970	nan	0.8408	0.9875	0.9936
0.0065	420.0	1680	0.0216	0.9413	0.9656	0.9938	nan	0.9052	0.9947	0.9969	nan	0.8424	0.9877	0.9937
0.0075	425.0	1700	0.0216	0.9417	0.9650	0.9939	nan	0.9028	0.9952	0.9969	nan	0.8434	0.9881	0.9938
0.0065	430.0	1720	0.0222	0.9404	0.9662	0.9937	nan	0.9076	0.9941	0.9969	nan	0.8401	0.9877	0.9935
0.008	435.0	1740	0.0226	0.9396	0.9662	0.9936	nan	0.9076	0.9945	0.9966	nan	0.8374	0.9879	0.9934
0.0083	440.0	1760	0.0219	0.9421	0.9659	0.9939	nan	0.9058	0.9949	0.9969	nan	0.8446	0.9879	0.9938
0.0072	445.0	1780	0.0220	0.9412	0.9666	0.9938	nan	0.9088	0.9942	0.9969	nan	0.8424	0.9875	0.9937
0.0074	450.0	1800	0.0224	0.9411	0.9647	0.9939	nan	0.9021	0.9952	0.9969	nan	0.8418	0.9878	0.9938
0.0069	455.0	1820	0.0219	0.9431	0.9662	0.9940	nan	0.9072	0.9944	0.9971	nan	0.8477	0.9877	0.9939
0.0072	460.0	1840	0.0221	0.9419	0.9665	0.9938	nan	0.9086	0.9939	0.9970	nan	0.8443	0.9877	0.9936
0.0078	465.0	1860	0.0222	0.9429	0.9646	0.9940	nan	0.9019	0.9945	0.9973	nan	0.8472	0.9877	0.9939
0.0072	470.0	1880	0.0230	0.9411	0.9655	0.9938	nan	0.9050	0.9946	0.9969	nan	0.8421	0.9876	0.9937
0.0076	475.0	1900	0.0223	0.9413	0.9657	0.9938	nan	0.9055	0.9946	0.9968	nan	0.8426	0.9877	0.9936
0.0072	480.0	1920	0.0221	0.9413	0.9665	0.9938	nan	0.9081	0.9947	0.9968	nan	0.8425	0.9879	0.9936
0.0073	485.0	1940	0.0225	0.9420	0.9664	0.9938	nan	0.9083	0.9936	0.9972	nan	0.8449	0.9875	0.9936
0.0079	490.0	1960	0.0226	0.9415	0.9664	0.9937	nan	0.9083	0.9939	0.9969	nan	0.8437	0.9875	0.9934
0.007	495.0	1980	0.0225	0.9415	0.9657	0.9937	nan	0.9059	0.9941	0.9970	nan	0.8434	0.9875	0.9935
0.0072	500.0	2000	0.0224	0.9416	0.9656	0.9938	nan	0.9055	0.9945	0.9969	nan	0.8435	0.9878	0.9935
0.0068	505.0	2020	0.0225	0.9413	0.9664	0.9938	nan	0.9081	0.9942	0.9969	nan	0.8428	0.9876	0.9936
0.0061	510.0	2040	0.0226	0.9431	0.9658	0.9939	nan	0.9065	0.9937	0.9973	nan	0.8481	0.9874	0.9937
0.0066	515.0	2060	0.0225	0.9419	0.9648	0.9939	nan	0.9021	0.9952	0.9969	nan	0.8441	0.9877	0.9937
0.0089	520.0	2080	0.0222	0.9433	0.9658	0.9939	nan	0.9059	0.9942	0.9972	nan	0.8486	0.9876	0.9937
0.0066	525.0	2100	0.0229	0.9415	0.9662	0.9938	nan	0.9075	0.9940	0.9970	nan	0.8434	0.9875	0.9936
0.0075	530.0	2120	0.0233	0.9402	0.9659	0.9937	nan	0.9068	0.9940	0.9969	nan	0.8396	0.9875	0.9935
0.0075	535.0	2140	0.0232	0.9407	0.9656	0.9938	nan	0.9054	0.9944	0.9969	nan	0.8409	0.9877	0.9936
0.007	540.0	2160	0.0229	0.9415	0.9662	0.9938	nan	0.9071	0.9946	0.9969	nan	0.8429	0.9880	0.9936
0.0071	545.0	2180	0.0224	0.9424	0.9658	0.9940	nan	0.9055	0.9949	0.9970	nan	0.8451	0.9883	0.9938
0.0069	550.0	2200	0.0225	0.9422	0.9663	0.9939	nan	0.9076	0.9941	0.9971	nan	0.8452	0.9878	0.9937
0.0067	555.0	2220	0.0223	0.9414	0.9661	0.9938	nan	0.9069	0.9946	0.9969	nan	0.8427	0.9880	0.9936
0.0066	560.0	2240	0.0227	0.9420	0.9659	0.9939	nan	0.9058	0.9948	0.9969	nan	0.8441	0.9880	0.9937
0.0069	565.0	2260	0.0230	0.9410	0.9660	0.9938	nan	0.9066	0.9945	0.9968	nan	0.8416	0.9878	0.9936
0.0065	570.0	2280	0.0227	0.9417	0.9661	0.9939	nan	0.9065	0.9949	0.9969	nan	0.8433	0.9881	0.9937
0.0069	575.0	2300	0.0226	0.9424	0.9664	0.9939	nan	0.9076	0.9944	0.9970	nan	0.8455	0.9879	0.9937
0.0065	580.0	2320	0.0227	0.9423	0.9661	0.9939	nan	0.9069	0.9945	0.9970	nan	0.8453	0.9880	0.9937
0.0069	585.0	2340	0.0226	0.9422	0.9661	0.9939	nan	0.9066	0.9947	0.9970	nan	0.8449	0.9880	0.9937
0.0062	590.0	2360	0.0226	0.9422	0.9661	0.9939	nan	0.9066	0.9945	0.9970	nan	0.8449	0.9879	0.9937
0.0068	595.0	2380	0.0227	0.9418	0.9660	0.9939	nan	0.9066	0.9944	0.9970	nan	0.8439	0.9879	0.9937
0.0073	600.0	2400	0.0226	0.9423	0.9658	0.9939	nan	0.9057	0.9946	0.9971	nan	0.8453	0.9880	0.9938

Framework versions

Transformers 4.30.2
Pytorch 2.0.1+cu117
Datasets 2.13.1
Tokenizers 0.13.3

Task: Image Segmentation
Library: transformers
Type: segformer
License: other
Created: Feb 25, 2024
Updated: Feb 26, 2024