You can deploy this model and start earning money today!

Vrjb/
SegFormer_b5_11

Name: SegFormer_b5_11
Availability: InStock

transformers

safetensors

segformer

generated_from_trainer

nvidia/segformer-b5-finetuned-cityscapes-1024-1024

finetune

other

endpoints_compatible

Views

No views yet

SegFormer_b5_11

This model is a fine-tuned version of nvidia/segformer-b5-finetuned-cityscapes-1024-1024 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.8010
Mean Iou: 0.7954
Mean Accuracy: 0.8798
Overall Accuracy: 0.9636
Accuracy Road: 0.9913
Accuracy Sidewalk: 0.9403
Accuracy Building: 0.9629
Accuracy Wall: 0.7410
Accuracy Fence: 0.7431
Accuracy Pole: 0.7549
Accuracy Traffic light: 0.8752
Accuracy Traffic sign: 0.9026
Accuracy Vegetation: 0.9649
Accuracy Terrain: 0.7458
Accuracy Sky: 0.9823
Accuracy Person: 0.9127
Accuracy Rider: 0.7910
Accuracy Car: 0.9800
Accuracy Truck: 0.8932
Accuracy Bus: 0.9593
Accuracy Train: 0.8554
Accuracy Motorcycle: 0.8280
Accuracy Bicycle: 0.8927
Iou Road: 0.9853
Iou Sidewalk: 0.8769
Iou Building: 0.9294
Iou Wall: 0.6725
Iou Fence: 0.6481
Iou Pole: 0.6122
Iou Traffic light: 0.7011
Iou Traffic sign: 0.8023
Iou Vegetation: 0.9259
Iou Terrain: 0.6490
Iou Sky: 0.9504
Iou Person: 0.8085
Iou Rider: 0.6295
Iou Car: 0.9521
Iou Truck: 0.8505
Iou Bus: 0.9051
Iou Train: 0.7672
Iou Motorcycle: 0.6727
Iou Bicycle: 0.7743

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: 9e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
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
lr_scheduler_warmup_steps: 1000
num_epochs: 20
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Accuracy Bicycle	Accuracy Building	Accuracy Bus	Accuracy Car	Accuracy Fence	Accuracy Motorcycle	Accuracy Person	Accuracy Pole	Accuracy Rider	Accuracy Road	Accuracy Sidewalk	Accuracy Sky	Accuracy Terrain	Accuracy Traffic light	Accuracy Traffic sign	Accuracy Train	Accuracy Truck	Accuracy Vegetation	Accuracy Wall	Iou Bicycle	Iou Building	Iou Bus	Iou Car	Iou Fence	Iou Motorcycle	Iou Person	Iou Pole	Iou Rider	Iou Road	Iou Sidewalk	Iou Sky	Iou Terrain	Iou Traffic light	Iou Traffic sign	Iou Train	Iou Truck	Iou Vegetation	Iou Wall	Validation Loss	Mean Accuracy	Mean Iou	Overall Accuracy
0.8456	0.5376	100	0.8747	0.9607	0.9437	0.9775	0.7085	0.7599	0.9048	0.7455	0.7610	0.9906	0.9288	0.9840	0.7789	0.8427	0.8786	0.8887	0.9006	0.9608	0.6635	0.7535	0.9230	0.8894	0.9483	0.5970	0.6374	0.7836	0.5859	0.5740	0.9834	0.8684	0.9456	0.6764	0.6691	0.7648	0.7901	0.8296	0.9231	0.5864	0.8201	0.8660	0.7752	0.9602
0.7465	1.0753	200	0.8953	0.9613	0.9506	0.9774	0.7230	0.8236	0.9016	0.7375	0.7563	0.9918	0.9356	0.9811	0.7888	0.8609	0.8764	0.8912	0.8998	0.9618	0.6790	0.7569	0.9234	0.8931	0.9516	0.6045	0.6414	0.7954	0.5904	0.5894	0.9854	0.8754	0.9497	0.6775	0.6721	0.7699	0.7690	0.8615	0.9251	0.5979	0.8066	0.8733	0.7805	0.9615
0.8764	1.6129	300	0.9008	0.9612	0.9433	0.9793	0.7547	0.8097	0.9005	0.7511	0.7752	0.9903	0.9396	0.9839	0.7864	0.8408	0.8819	0.9089	0.9136	0.9622	0.6706	0.7598	0.9246	0.8930	0.9510	0.6305	0.6610	0.7979	0.5977	0.6033	0.9851	0.8752	0.9470	0.6769	0.6801	0.7820	0.7666	0.8635	0.9255	0.6064	0.8097	0.8765	0.7856	0.9620
0.7359	2.1505	400	0.8901	0.9607	0.9386	0.9761	0.7167	0.8253	0.9077	0.7430	0.7825	0.9905	0.9391	0.9838	0.7778	0.8542	0.8748	0.8920	0.9048	0.9661	0.6947	0.7636	0.9245	0.8868	0.9512	0.6363	0.6503	0.7985	0.6001	0.6098	0.9845	0.8719	0.9484	0.6744	0.6749	0.7769	0.7577	0.8573	0.9253	0.6225	0.8090	0.8747	0.7850	0.9620
0.7919	2.6882	500	0.9014	0.9609	0.9466	0.9799	0.7401	0.8459	0.9047	0.7518	0.7771	0.9908	0.9401	0.9845	0.7713	0.8783	0.8932	0.8925	0.8930	0.9611	0.7098	0.7544	0.9257	0.8922	0.9507	0.6480	0.6348	0.8013	0.6025	0.6135	0.9848	0.8734	0.9477	0.6692	0.6717	0.7750	0.7866	0.8403	0.9260	0.6263	0.8023	0.8802	0.7855	0.9622
0.7191	3.2258	600	0.8953	0.9579	0.9395	0.9787	0.7650	0.8355	0.9048	0.7311	0.7800	0.9899	0.9404	0.9851	0.7621	0.8686	0.8920	0.8988	0.8892	0.9679	0.6577	0.7628	0.9246	0.8904	0.9507	0.6388	0.6582	0.7989	0.6031	0.6143	0.9843	0.8700	0.9461	0.6666	0.6777	0.7833	0.7473	0.8522	0.9246	0.5967	0.8065	0.8758	0.7837	0.9617
0.7745	3.7634	700	0.8943	0.9637	0.9493	0.9785	0.7420	0.8280	0.9097	0.7325	0.8065	0.9904	0.9360	0.9842	0.7625	0.8534	0.8953	0.8598	0.9001	0.9622	0.7256	0.7607	0.9268	0.8841	0.9506	0.6273	0.6433	0.7975	0.6022	0.6199	0.9847	0.8724	0.9478	0.6619	0.6930	0.7912	0.7575	0.8396	0.9258	0.6443	0.8021	0.8776	0.7858	0.9623
0.7367	4.3011	800	0.9093	0.9618	0.9449	0.9781	0.7500	0.8337	0.9044	0.7514	0.7928	0.9901	0.9353	0.9827	0.7850	0.8725	0.8896	0.8837	0.8948	0.9617	0.7004	0.7460	0.9268	0.8929	0.9504	0.6400	0.6321	0.8002	0.6050	0.6170	0.9840	0.8686	0.9485	0.6595	0.6862	0.7912	0.7854	0.8475	0.9246	0.6433	0.7997	0.8801	0.7868	0.9619
0.7503	4.8387	900	0.8817	0.9648	0.9511	0.9766	0.7600	0.8513	0.9242	0.7315	0.7496	0.9902	0.9400	0.9846	0.7469	0.8654	0.8977	0.8548	0.8720	0.9612	0.7353	0.7647	0.9296	0.8866	0.9496	0.6262	0.6102	0.7943	0.6053	0.6080	0.9846	0.8724	0.9470	0.6628	0.6888	0.7974	0.7076	0.8258	0.9255	0.6435	0.7955	0.8757	0.7805	0.9625
0.6957	5.3763	1000	0.8892	0.9658	0.9637	0.9769	0.7331	0.8256	0.9122	0.7369	0.7667	0.9905	0.9398	0.9807	0.7519	0.8553	0.8957	0.8498	0.8989	0.9613	0.7224	0.7631	0.9276	0.8939	0.9507	0.6321	0.6350	0.7996	0.6024	0.6125	0.9845	0.8712	0.9512	0.6634	0.6916	0.7959	0.6943	0.8423	0.9257	0.6415	0.7978	0.8746	0.7831	0.9625
0.7057	5.9140	1100	0.8826	0.9633	0.9504	0.9792	0.7497	0.7977	0.9110	0.7324	0.8061	0.9915	0.9366	0.9827	0.7486	0.8732	0.8910	0.8817	0.8972	0.9635	0.7321	0.7690	0.9281	0.8900	0.9509	0.6408	0.6762	0.8004	0.6055	0.6196	0.9849	0.8746	0.9519	0.6451	0.6913	0.7928	0.7596	0.8476	0.9252	0.6504	0.7907	0.8774	0.7897	0.9628
0.7232	6.4516	1200	0.9046	0.9633	0.9497	0.9790	0.7542	0.8296	0.9038	0.7324	0.7938	0.9907	0.9399	0.9861	0.7605	0.8725	0.9006	0.8611	0.8941	0.9620	0.7467	0.7650	0.9290	0.8919	0.9512	0.6442	0.6748	0.8036	0.6041	0.6233	0.9850	0.8746	0.9474	0.6523	0.6870	0.7915	0.7657	0.8481	0.9249	0.6596	0.7927	0.8802	0.7907	0.9629
0.7168	6.9892	1300	0.8830	0.9650	0.9548	0.9790	0.7742	0.8130	0.9006	0.7287	0.8121	0.9913	0.9380	0.9831	0.7694	0.8789	0.8839	0.8345	0.9056	0.9628	0.7568	0.7677	0.9305	0.8870	0.9509	0.6559	0.6532	0.8054	0.6054	0.6168	0.9851	0.8745	0.9513	0.6673	0.6920	0.7904	0.7415	0.8340	0.9259	0.6668	0.7881	0.8797	0.7896	0.9634
0.8144	7.5269	1400	0.8928	0.9669	0.9577	0.9762	0.7470	0.8043	0.9186	0.7260	0.7572	0.9909	0.9455	0.9835	0.7529	0.8334	0.8796	0.8693	0.9011	0.9599	0.7138	0.7656	0.9277	0.8981	0.9519	0.6414	0.6424	0.8027	0.6054	0.6195	0.9852	0.8744	0.9491	0.6484	0.6973	0.7954	0.7597	0.8504	0.9256	0.6275	0.7988	0.8724	0.7878	0.9628
0.674	8.0645	1500	0.8897	0.9579	0.9578	0.9795	0.7540	0.8216	0.9106	0.7614	0.8117	0.9894	0.9419	0.9789	0.7606	0.8837	0.9095	0.8278	0.9047	0.9660	0.7330	0.7675	0.9271	0.8821	0.9497	0.6487	0.6595	0.8033	0.6039	0.6307	0.9845	0.8710	0.9505	0.6458	0.6866	0.7881	0.7477	0.8369	0.9245	0.6700	0.7934	0.8810	0.7883	0.9623
0.7338	8.6022	1600	0.9060	0.9592	0.9399	0.9803	0.7338	0.8360	0.9070	0.7568	0.7956	0.9905	0.9415	0.9804	0.7382	0.8807	0.9113	0.8447	0.8667	0.9657	0.7684	0.7642	0.9283	0.8809	0.9501	0.6497	0.6348	0.8044	0.6017	0.6277	0.9852	0.8749	0.9497	0.6511	0.6893	0.7935	0.7579	0.8287	0.9252	0.6553	0.8012	0.8791	0.7870	0.9627
0.732	9.1398	1700	0.9015	0.9619	0.9567	0.9780	0.7595	0.7503	0.9102	0.7426	0.8075	0.9915	0.9402	0.9811	0.7447	0.8711	0.8977	0.8600	0.8946	0.9651	0.7170	0.7647	0.9284	0.8980	0.9522	0.6496	0.6533	0.8081	0.6072	0.6366	0.9856	0.8772	0.9515	0.6495	0.6977	0.7988	0.7542	0.8434	0.9246	0.6266	0.7934	0.8753	0.7899	0.9631
0.7185	9.6774	1800	0.8829	0.9621	0.9505	0.9795	0.7833	0.8020	0.9157	0.7461	0.7981	0.9909	0.9382	0.9832	0.7786	0.8596	0.8963	0.8676	0.8832	0.9603	0.7716	0.7691	0.9299	0.8980	0.9516	0.6466	0.6670	0.8060	0.6063	0.6225	0.9849	0.8739	0.9499	0.6512	0.6990	0.7985	0.7668	0.8446	0.9247	0.6516	0.7952	0.8816	0.7917	0.9629
0.7091	10.2151	1900	0.8912	0.9648	0.9520	0.9794	0.7253	0.8268	0.9211	0.7488	0.7938	0.9911	0.9429	0.9824	0.7679	0.8810	0.8992	0.8435	0.8940	0.9610	0.7557	0.7716	0.9294	0.8991	0.9525	0.6459	0.6698	0.8056	0.6074	0.6445	0.9854	0.8761	0.9509	0.6601	0.6914	0.7984	0.7805	0.8535	0.9260	0.6588	0.7921	0.8801	0.7951	0.9635
0.6826	10.7527	2000	0.8785	0.9637	0.9457	0.9801	0.6996	0.8356	0.9104	0.7530	0.8009	0.9906	0.9428	0.9805	0.7494	0.8558	0.8981	0.8121	0.8620	0.9660	0.7376	0.7694	0.9284	0.8812	0.9508	0.6287	0.6513	0.8094	0.6060	0.6381	0.9849	0.8741	0.9528	0.6593	0.7036	0.7983	0.7298	0.8311	0.9256	0.6666	0.7996	0.8717	0.7889	0.9631
0.7622	11.2903	2100	0.8916	0.9630	0.9596	0.9799	0.7587	0.7991	0.9095	0.7388	0.8078	0.9907	0.9427	0.9823	0.7546	0.8863	0.8897	0.8390	0.8936	0.9643	0.7664	0.7684	0.9306	0.8943	0.9520	0.6393	0.6828	0.8097	0.6067	0.6298	0.9851	0.8743	0.9516	0.6589	0.6909	0.7975	0.7664	0.8493	0.9261	0.6653	0.7964	0.8799	0.7936	0.9634
0.7119	11.8280	2200	0.8803	0.9621	0.9525	0.9806	0.7773	0.7930	0.9044	0.7455	0.8255	0.9906	0.9395	0.9807	0.7635	0.8761	0.8934	0.8577	0.8865	0.9642	0.7259	0.7688	0.9300	0.8945	0.9511	0.6536	0.6718	0.8069	0.6057	0.6263	0.9848	0.8726	0.9509	0.6453	0.6942	0.7977	0.7797	0.8432	0.9247	0.6438	0.7983	0.8789	0.7919	0.9629
0.7236	12.3656	2300	0.8938	0.9609	0.9551	0.9794	0.7626	0.8259	0.9156	0.7575	0.7791	0.9907	0.9408	0.9813	0.7429	0.8790	0.9012	0.8580	0.8974	0.9643	0.7481	0.7711	0.9289	0.8987	0.9510	0.6524	0.6577	0.8062	0.6050	0.6293	0.9848	0.8727	0.9500	0.6531	0.6968	0.8004	0.7965	0.8465	0.9252	0.6718	0.7972	0.8807	0.7946	0.9631
0.745	12.9032	2400	0.8928	0.9615	0.9572	0.9790	0.7629	0.8120	0.9063	0.7594	0.8244	0.9913	0.9409	0.9826	0.7544	0.8728	0.8969	0.8665	0.8985	0.9648	0.7048	0.7716	0.9300	0.8993	0.9510	0.6337	0.6622	0.8115	0.6105	0.6402	0.9854	0.8765	0.9505	0.6522	0.7016	0.7979	0.7610	0.8434	0.9258	0.6376	0.7979	0.8805	0.7917	0.9633
0.7197	13.4409	2500	0.8929	0.9647	0.9573	0.9787	0.7667	0.8368	0.9116	0.7487	0.8204	0.9919	0.9397	0.9835	0.7415	0.8728	0.8990	0.8259	0.8953	0.9609	0.7300	0.7751	0.9307	0.8883	0.9520	0.6490	0.6705	0.8093	0.6107	0.6402	0.9853	0.8744	0.9496	0.6444	0.6988	0.8019	0.7652	0.8506	0.9256	0.6509	0.7964	0.8799	0.7933	0.9635
0.6799	13.9785	2600	0.8919	0.9614	0.9597	0.9801	0.7438	0.8240	0.9139	0.7602	0.8042	0.9915	0.9367	0.9835	0.7437	0.8714	0.8954	0.8393	0.8935	0.9642	0.7542	0.7718	0.9294	0.8916	0.9516	0.6441	0.6707	0.8078	0.6099	0.6347	0.9848	0.8734	0.9496	0.6567	0.6986	0.7943	0.7562	0.8487	0.9260	0.6703	0.7988	0.8796	0.7932	0.9632
0.7878	14.5161	2700	0.8927	0.9620	0.9583	0.9793	0.7536	0.8011	0.9151	0.7567	0.7931	0.9914	0.9375	0.9821	0.7587	0.8762	0.8942	0.8430	0.8929	0.9650	0.7350	0.7745	0.9292	0.8971	0.9516	0.6475	0.6774	0.8057	0.6080	0.6251	0.9852	0.8776	0.9497	0.6536	0.6989	0.8010	0.7671	0.8465	0.9258	0.6661	0.7984	0.8783	0.7941	0.9634
0.7148	15.0538	2800	0.8931	0.9628	0.9587	0.9797	0.7453	0.8243	0.9091	0.7500	0.7917	0.9916	0.9399	0.9812	0.7571	0.8768	0.8993	0.8510	0.8947	0.9654	0.7297	0.7730	0.9296	0.9011	0.9522	0.6505	0.6674	0.8061	0.6091	0.6236	0.9853	0.8778	0.9504	0.6558	0.7039	0.7955	0.7517	0.8510	0.9259	0.6642	0.7981	0.8790	0.7934	0.9636
0.7285	15.5914	2900	0.8943	0.9606	0.9532	0.9801	0.7467	0.8390	0.9101	0.7498	0.7744	0.9914	0.9405	0.9805	0.7476	0.8748	0.8958	0.8665	0.8940	0.9672	0.7377	0.7741	0.9296	0.9035	0.9517	0.6470	0.6514	0.8072	0.6080	0.6198	0.9852	0.8764	0.9519	0.6497	0.7023	0.7972	0.7651	0.8472	0.9251	0.6624	0.7999	0.8792	0.7924	0.9634
0.7323	16.1290	3000	0.8910	0.9625	0.9586	0.9793	0.7261	0.8325	0.9114	0.7610	0.7587	0.9920	0.9380	0.9820	0.7519	0.8684	0.8963	0.8765	0.8963	0.9653	0.7514	0.7733	0.9293	0.9072	0.9525	0.6437	0.6599	0.8056	0.6113	0.6173	0.9856	0.8784	0.9510	0.6500	0.7041	0.8001	0.7632	0.8499	0.9257	0.6764	0.8014	0.8789	0.7939	0.9636
0.7061	16.6667	3100	0.8921	0.9628	0.9597	0.9798	0.7371	0.8213	0.9142	0.7492	0.7705	0.9918	0.9378	0.9825	0.7506	0.8748	0.8998	0.8767	0.8934	0.9655	0.7478	0.7745	0.9296	0.9068	0.9519	0.6456	0.6716	0.8063	0.6108	0.6240	0.9854	0.8777	0.9500	0.6536	0.6998	0.8011	0.7769	0.8482	0.9259	0.6744	0.7985	0.8793	0.7955	0.9637
0.6841	17.2043	3200	0.8868	0.9633	0.9550	0.9794	0.7299	0.8301	0.9135	0.7500	0.7975	0.9912	0.9429	0.9835	0.7561	0.8647	0.9024	0.8522	0.8900	0.9640	0.7625	0.7730	0.9300	0.9052	0.9521	0.6484	0.6691	0.8070	0.6104	0.6292	0.9854	0.8775	0.9497	0.6519	0.7022	0.8053	0.7726	0.8495	0.9257	0.6717	0.8015	0.8797	0.7956	0.9637
0.6523	17.7419	3300	0.8919	0.9617	0.9595	0.9802	0.7629	0.8459	0.9139	0.7584	0.7935	0.9921	0.9339	0.9828	0.7501	0.8744	0.8993	0.8529	0.8921	0.9652	0.7499	0.7733	0.9304	0.9021	0.9519	0.6569	0.6581	0.8088	0.6126	0.6330	0.9853	0.8772	0.9501	0.6491	0.7004	0.8028	0.7517	0.8487	0.9258	0.6701	0.7975	0.8821	0.7941	0.9637
0.723	18.2796	3400	0.8018	0.7947	0.8804	0.9635	0.9910	0.9418	0.9621	0.7371	0.7366	0.7561	0.8741	0.9049	0.9654	0.7434	0.9822	0.9172	0.7981	0.9795	0.8968	0.9583	0.8561	0.8354	0.8917	0.9853	0.8765	0.9293	0.6625	0.6469	0.6115	0.6992	0.8035	0.9258	0.6473	0.9504	0.8072	0.6330	0.9520	0.8479	0.9057	0.7689	0.6735	0.7737
0.6937	18.8172	3500	0.8023	0.7949	0.8793	0.9636	0.9916	0.9381	0.9636	0.7394	0.7354	0.7518	0.8733	0.9042	0.9642	0.7494	0.9831	0.9147	0.7860	0.9800	0.8926	0.9595	0.8541	0.8311	0.8939	0.9854	0.8770	0.9293	0.6726	0.6458	0.6119	0.6998	0.8048	0.9260	0.6485	0.9500	0.8073	0.6269	0.9521	0.8494	0.9042	0.7645	0.6728	0.7739
0.7394	19.3548	3600	0.8042	0.7947	0.8784	0.9635	0.9914	0.9395	0.9634	0.7360	0.7353	0.7553	0.8686	0.8979	0.9643	0.7476	0.9831	0.9135	0.7847	0.9800	0.8929	0.9589	0.8539	0.8324	0.8906	0.9853	0.8768	0.9290	0.6710	0.6472	0.6118	0.7025	0.8008	0.9258	0.6498	0.9500	0.8076	0.6266	0.9520	0.8499	0.9052	0.7664	0.6679	0.7742
0.7267	19.8925	3700	0.8010	0.7954	0.8798	0.9636	0.9913	0.9403	0.9629	0.7410	0.7431	0.7549	0.8752	0.9026	0.9649	0.7458	0.9823	0.9127	0.7910	0.9800	0.8932	0.9593	0.8554	0.8280	0.8927	0.9853	0.8769	0.9294	0.6725	0.6481	0.6122	0.7011	0.8023	0.9259	0.6490	0.9504	0.8085	0.6295	0.9521	0.8505	0.9051	0.7672	0.6727	0.7743

Framework versions

Transformers 4.48.3
Pytorch 2.1.2+cu121
Datasets 3.2.0
Tokenizers 0.21.0

Vrjb/SegFormer_b5_11

SegFormer_b5_11

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Vrjb/
SegFormer_b5_11