segformer-b0-finetuned-segments-GFB

luoyun75579/segformer-b0-finetuned-segments-GFB

segformer-b0-finetuned-segments-GFB is a image segmentation model from luoyun75579. 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

9% of all-time downloads

All-time downloads

Public

Parameters

3.7M

14.9 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors14.9 MB · 100%

At a glance

Task: Image Segmentation
Library: transformers
License: other
Model type: segformer
Access: Public
Created: Nov 15, 2025
Updated: Nov 15, 2025
SHA: 5891e4a2

Base models

nvidia/mit-b0

Model card

From the Hugging Face model README

segformer-b0-finetuned-segments-GFB

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

Loss: 0.3870
Mean Iou: 0.6847
Mean Accuracy: 0.7892
Overall Accuracy: 0.9154
Accuracy Unlabeled: 0.9601
Accuracy Gbm: 0.8096
Accuracy Podo: 0.7494
Accuracy Endo: 0.6375
Iou Unlabeled: 0.9135
Iou Gbm: 0.6883
Iou Podo: 0.6138
Iou Endo: 0.5232

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: 0.0005
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 10
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Gbm	Accuracy Podo	Accuracy Endo	Iou Unlabeled	Iou Gbm	Iou Podo	Iou Endo
0.5197	0.5495	100	0.4048	0.4345	0.5150	0.8510	0.9671	0.5512	0.4744	0.0673	0.8616	0.4810	0.3306	0.0647
0.248	1.0989	200	0.3859	0.5298	0.6920	0.8453	0.8921	0.6956	0.7774	0.4028	0.8483	0.5528	0.4084	0.3097
0.2993	1.6484	300	0.3585	0.4983	0.5661	0.8670	0.9790	0.6002	0.4155	0.2698	0.8669	0.5140	0.3661	0.2461
0.2063	2.1978	400	0.2966	0.5869	0.6720	0.8893	0.9703	0.6824	0.5846	0.4506	0.8884	0.5880	0.4947	0.3765
0.3007	2.7473	500	0.3307	0.5586	0.6353	0.8834	0.9746	0.6545	0.5395	0.3727	0.8826	0.5729	0.4572	0.3216
0.2984	3.2967	600	0.2688	0.6161	0.7325	0.8954	0.9566	0.7541	0.6391	0.5804	0.8979	0.6246	0.5209	0.4212
0.1996	3.8462	700	0.2812	0.6039	0.6978	0.8910	0.9650	0.6807	0.6325	0.5131	0.8880	0.6008	0.5136	0.4130
0.2482	4.3956	800	0.2931	0.6014	0.6822	0.8946	0.9722	0.6933	0.6148	0.4485	0.8931	0.6080	0.5131	0.3913
0.533	4.9451	900	0.2568	0.6299	0.7261	0.9018	0.9653	0.7396	0.6741	0.5252	0.9025	0.6396	0.5424	0.4350
0.1977	5.4945	1000	0.2651	0.6350	0.7845	0.8957	0.9332	0.8315	0.7323	0.6411	0.8965	0.6483	0.5557	0.4396
0.2118	6.0440	1100	0.2630	0.6340	0.7449	0.9019	0.9587	0.7901	0.6482	0.5826	0.9033	0.6509	0.5404	0.4414
0.221	6.5934	1200	0.2697	0.6451	0.7482	0.9055	0.9673	0.7277	0.6825	0.6151	0.9051	0.6464	0.5682	0.4605
0.189	7.1429	1300	0.2537	0.6460	0.7543	0.9051	0.9538	0.7805	0.7689	0.5142	0.9043	0.6672	0.5783	0.4341
0.2167	7.6923	1400	0.2687	0.6431	0.7999	0.8969	0.9284	0.8781	0.7140	0.6790	0.8979	0.6467	0.5639	0.4641
0.1821	8.2418	1500	0.2824	0.6386	0.7578	0.8976	0.9481	0.7569	0.7340	0.5923	0.8956	0.6318	0.5565	0.4705
0.1373	8.7912	1600	0.2710	0.6447	0.7554	0.9042	0.9548	0.8047	0.6983	0.5638	0.9039	0.6586	0.5674	0.4489
0.1611	9.3407	1700	0.2825	0.6499	0.7580	0.9041	0.9609	0.7409	0.7024	0.6278	0.9023	0.6448	0.5701	0.4822
0.1751	9.8901	1800	0.3101	0.6321	0.7211	0.9001	0.9702	0.7075	0.6343	0.5723	0.8969	0.6200	0.5438	0.4677
0.1199	10.4396	1900	0.2695	0.6648	0.7891	0.9070	0.9501	0.7935	0.7543	0.6583	0.9052	0.6660	0.5921	0.4957
0.1328	10.9890	2000	0.2586	0.6600	0.7992	0.9038	0.9363	0.8560	0.7674	0.6371	0.9025	0.6781	0.5755	0.4840
0.1444	11.5385	2100	0.2983	0.6428	0.7713	0.9029	0.9569	0.7414	0.7048	0.6820	0.9043	0.6489	0.5694	0.4484
0.204	12.0879	2200	0.2692	0.6597	0.8073	0.9031	0.9358	0.8507	0.7516	0.6912	0.9019	0.6695	0.5841	0.4831
0.1552	12.6374	2300	0.2682	0.6469	0.7417	0.9067	0.9622	0.7762	0.7163	0.5123	0.9056	0.6643	0.5701	0.4476
0.1459	13.1868	2400	0.2931	0.6491	0.7420	0.9079	0.9668	0.7597	0.7000	0.5416	0.9076	0.6632	0.5679	0.4576
0.2669	13.7363	2500	0.2840	0.6481	0.7535	0.9062	0.9563	0.8487	0.6537	0.5555	0.9067	0.6639	0.5595	0.4623
0.09	14.2857	2600	0.3038	0.6689	0.7742	0.9098	0.9616	0.7601	0.7318	0.6432	0.9071	0.6677	0.5910	0.5100
0.1161	14.8352	2700	0.2944	0.6588	0.7654	0.9081	0.9653	0.7399	0.7009	0.6554	0.9073	0.6589	0.5737	0.4953
0.1315	15.3846	2800	0.2982	0.6708	0.7997	0.9089	0.9522	0.7881	0.7444	0.7139	0.9076	0.6695	0.5974	0.5087
0.1709	15.9341	2900	0.2805	0.6720	0.7900	0.9104	0.9539	0.8161	0.7302	0.6596	0.9091	0.6769	0.5950	0.5070
0.1406	16.4835	3000	0.2955	0.6664	0.7778	0.9093	0.9533	0.8202	0.7400	0.5977	0.9080	0.6728	0.5930	0.4917
0.102	17.0330	3100	0.3193	0.6602	0.7738	0.9073	0.9553	0.7973	0.7171	0.6257	0.9053	0.6667	0.5852	0.4836
0.135	17.5824	3200	0.2937	0.6683	0.7766	0.9100	0.9568	0.8114	0.7197	0.6186	0.9082	0.6781	0.5857	0.5013
0.1284	18.1319	3300	0.2968	0.6695	0.7921	0.9079	0.9468	0.8185	0.7727	0.6304	0.9066	0.6722	0.5907	0.5084
0.0931	18.6813	3400	0.2802	0.6729	0.7810	0.9127	0.9583	0.7982	0.7583	0.6093	0.9124	0.6782	0.6037	0.4975
0.1284	19.2308	3500	0.3265	0.6656	0.7606	0.9113	0.9670	0.7696	0.7019	0.6038	0.9104	0.6663	0.5849	0.5007
0.1034	19.7802	3600	0.2927	0.6748	0.7903	0.9114	0.9525	0.8156	0.7677	0.6255	0.9097	0.6801	0.6064	0.5028
0.11	20.3297	3700	0.2977	0.6770	0.8022	0.9115	0.9482	0.8299	0.7795	0.6512	0.9103	0.6837	0.6112	0.5027
0.1036	20.8791	3800	0.3123	0.6736	0.7843	0.9112	0.9579	0.7919	0.7385	0.6487	0.9088	0.6759	0.6013	0.5086
0.1069	21.4286	3900	0.3274	0.6729	0.7683	0.9136	0.9658	0.7858	0.7208	0.6008	0.9121	0.6808	0.5981	0.5005
0.1383	21.9780	4000	0.3168	0.6789	0.7906	0.9133	0.9563	0.8249	0.7361	0.6452	0.9124	0.6826	0.6039	0.5167
0.0709	22.5275	4100	0.3164	0.6834	0.7998	0.9136	0.9555	0.8177	0.7404	0.6857	0.9122	0.6835	0.6058	0.5320
0.1165	23.0769	4200	0.3564	0.6730	0.7790	0.9117	0.9627	0.7539	0.7517	0.6479	0.9097	0.6695	0.6032	0.5098
0.1421	23.6264	4300	0.3332	0.6742	0.7722	0.9128	0.9645	0.7797	0.7229	0.6216	0.9110	0.6747	0.5958	0.5151
0.082	24.1758	4400	0.3304	0.6723	0.7866	0.9098	0.9507	0.8117	0.7764	0.6076	0.9074	0.6762	0.6024	0.5030
0.0852	24.7253	4500	0.3538	0.6736	0.7830	0.9114	0.9589	0.8033	0.7149	0.6548	0.9094	0.6774	0.5939	0.5137
0.1127	25.2747	4600	0.3505	0.6711	0.7663	0.9125	0.9652	0.7862	0.7122	0.6014	0.9103	0.6749	0.5974	0.5019
0.077	25.8242	4700	0.3504	0.6704	0.7619	0.9125	0.9669	0.7795	0.7096	0.5915	0.9100	0.6731	0.5971	0.5016
0.0875	26.3736	4800	0.3206	0.6799	0.7892	0.9137	0.9565	0.8298	0.7356	0.6349	0.9122	0.6856	0.6060	0.5160
0.1088	26.9231	4900	0.3283	0.6755	0.7821	0.9126	0.9546	0.8397	0.7428	0.5913	0.9105	0.6885	0.6021	0.5008
0.0786	27.4725	5000	0.3493	0.6758	0.7861	0.9118	0.9547	0.8146	0.7546	0.6207	0.9095	0.6846	0.6006	0.5085
0.0864	28.0220	5100	0.3285	0.6740	0.7738	0.9132	0.9600	0.8033	0.7539	0.5778	0.9108	0.6828	0.6100	0.4923
0.079	28.5714	5200	0.3304	0.6805	0.7860	0.9138	0.9589	0.8019	0.7546	0.6288	0.9116	0.6847	0.6084	0.5171
0.0631	29.1209	5300	0.3396	0.6757	0.7742	0.9138	0.9630	0.8154	0.7058	0.6127	0.9123	0.6822	0.5982	0.5100
0.1081	29.6703	5400	0.3386	0.6838	0.8016	0.9131	0.9536	0.8183	0.7536	0.6807	0.9112	0.6881	0.6033	0.5327
0.0673	30.2198	5500	0.3377	0.6800	0.7931	0.9131	0.9571	0.8061	0.7423	0.6667	0.9112	0.6814	0.6075	0.5197
0.0675	30.7692	5600	0.3428	0.6816	0.7817	0.9152	0.9630	0.8067	0.7252	0.6318	0.9133	0.6878	0.6056	0.5197
0.157	31.3187	5700	0.3372	0.6812	0.7843	0.9144	0.9598	0.8128	0.7407	0.6239	0.9126	0.6851	0.6082	0.5191
0.0927	31.8681	5800	0.3470	0.6789	0.7880	0.9131	0.9570	0.8190	0.7367	0.6391	0.9110	0.6835	0.6056	0.5156
0.0498	32.4176	5900	0.3469	0.6794	0.7834	0.9142	0.9588	0.8220	0.7383	0.6147	0.9126	0.6865	0.6057	0.5128
0.1188	32.9670	6000	0.3384	0.6858	0.8022	0.9140	0.9518	0.8327	0.7717	0.6526	0.9122	0.6898	0.6120	0.5292
0.046	33.5165	6100	0.3590	0.6776	0.7820	0.9132	0.9589	0.8024	0.7510	0.6157	0.9110	0.6830	0.6062	0.5103
0.0915	34.0659	6200	0.3670	0.6810	0.7831	0.9147	0.9621	0.7928	0.7476	0.6297	0.9127	0.6831	0.6116	0.5167
0.0455	34.6154	6300	0.3617	0.6797	0.7871	0.9131	0.9578	0.8043	0.7509	0.6355	0.9106	0.6822	0.6092	0.5170
0.0489	35.1648	6400	0.3655	0.6819	0.7854	0.9144	0.9603	0.7975	0.7551	0.6289	0.9121	0.6830	0.6139	0.5186
0.0683	35.7143	6500	0.3830	0.6778	0.7771	0.9140	0.9639	0.7877	0.7296	0.6271	0.9117	0.6781	0.6089	0.5123
0.0779	36.2637	6600	0.3578	0.6862	0.7946	0.9152	0.9580	0.8122	0.7577	0.6505	0.9129	0.6889	0.6169	0.5262
0.1016	36.8132	6700	0.3546	0.6841	0.7908	0.9150	0.9573	0.8231	0.7567	0.6261	0.9132	0.6904	0.6130	0.5197
0.0695	37.3626	6800	0.3635	0.6855	0.7960	0.9150	0.9566	0.8155	0.7654	0.6467	0.9135	0.6877	0.6151	0.5258
0.0717	37.9121	6900	0.3656	0.6843	0.7896	0.9151	0.9586	0.8132	0.7559	0.6308	0.9130	0.6877	0.6146	0.5220
0.0805	38.4615	7000	0.3699	0.6851	0.7871	0.9158	0.9614	0.8051	0.7488	0.6330	0.9136	0.6882	0.6153	0.5234
0.0988	39.0110	7100	0.3761	0.6834	0.7854	0.9154	0.9618	0.8077	0.7366	0.6355	0.9135	0.6867	0.6104	0.5229
0.0802	39.5604	7200	0.3811	0.6823	0.7810	0.9154	0.9629	0.8032	0.7385	0.6195	0.9133	0.6866	0.6119	0.5176
0.0553	40.1099	7300	0.3718	0.6843	0.7875	0.9155	0.9600	0.8155	0.7470	0.6275	0.9139	0.6887	0.6115	0.5233
0.0549	40.6593	7400	0.3699	0.6858	0.7916	0.9158	0.9590	0.8205	0.7492	0.6375	0.9140	0.6912	0.6148	0.5230
0.0682	41.2088	7500	0.3750	0.6871	0.7957	0.9156	0.9579	0.8181	0.7546	0.6523	0.9136	0.6907	0.6167	0.5275
0.0831	41.7582	7600	0.3949	0.6809	0.7780	0.9149	0.9639	0.7921	0.7397	0.6164	0.9124	0.6818	0.6135	0.5161
0.072	42.3077	7700	0.3730	0.6859	0.7973	0.9148	0.9552	0.8264	0.7598	0.6479	0.9129	0.6886	0.6146	0.5276
0.0417	42.8571	7800	0.3853	0.6835	0.7857	0.9152	0.9609	0.8055	0.7483	0.6281	0.9131	0.6861	0.6138	0.5208
0.0438	43.4066	7900	0.3795	0.6844	0.7903	0.9150	0.9589	0.8126	0.7513	0.6386	0.9130	0.6881	0.6132	0.5235
0.076	43.9560	8000	0.3862	0.6836	0.7854	0.9155	0.9617	0.8074	0.7402	0.6322	0.9134	0.6882	0.6117	0.5212
0.0754	44.5055	8100	0.3854	0.6831	0.7847	0.9154	0.9614	0.8079	0.7432	0.6261	0.9134	0.6875	0.6124	0.5189
0.088	45.0549	8200	0.3976	0.6840	0.7868	0.9154	0.9616	0.8041	0.7423	0.6393	0.9135	0.6865	0.6126	0.5235
0.0468	45.6044	8300	0.3893	0.6841	0.7875	0.9154	0.9607	0.8084	0.7449	0.6360	0.9135	0.6876	0.6124	0.5229
0.0599	46.1538	8400	0.3837	0.6851	0.7891	0.9155	0.9604	0.8085	0.7487	0.6389	0.9135	0.6883	0.6139	0.5248
0.0869	46.7033	8500	0.3816	0.6850	0.7893	0.9155	0.9601	0.8100	0.7497	0.6372	0.9136	0.6889	0.6139	0.5237
0.0585	47.2527	8600	0.3876	0.6844	0.7879	0.9155	0.9604	0.8094	0.7498	0.6319	0.9135	0.6885	0.6140	0.5218
0.0754	47.8022	8700	0.3861	0.6834	0.7852	0.9155	0.9610	0.8133	0.7414	0.6253	0.9135	0.6888	0.6120	0.5194
0.0692	48.3516	8800	0.3863	0.6847	0.7890	0.9155	0.9601	0.8115	0.7478	0.6365	0.9136	0.6887	0.6136	0.5231
0.0486	48.9011	8900	0.3854	0.6844	0.7878	0.9155	0.9607	0.8091	0.7462	0.6352	0.9136	0.6883	0.6132	0.5226
0.0788	49.4505	9000	0.3925	0.6840	0.7869	0.9155	0.9610	0.8078	0.7456	0.6331	0.9135	0.6877	0.6130	0.5218
0.0746	50.0	9100	0.3870	0.6847	0.7892	0.9154	0.9601	0.8096	0.7494	0.6375	0.9135	0.6883	0.6138	0.5232

Framework versions

Transformers 4.57.1
Pytorch 2.9.1+cu130
Datasets 4.4.1
Tokenizers 0.22.1

Task: Image Segmentation
Library: transformers
Type: segformer
License: other
Created: Nov 15, 2025
Updated: Nov 15, 2025