You can deploy this model and start earning money today!

NICOPOI-9/
segformer-b0-finetuned-ade20k-morphpadver1-hgo-coord-v9_mix_resample_40epochs

Name: segformer-b0-finetuned-ade20k-morphpadver1-hgo-coord-v9_mix_resample_40epochs
Availability: InStock

generated_from_trainer

nvidia/segformer-b0-finetuned-ade-512-512

finetune

other

endpoints_compatible

Views

No views yet

segformer-b0-finetuned-ade20k-morphpadver1-hgo-coord-v9_mix_resample_40epochs

This model is a fine-tuned version of nvidia/segformer-b0-finetuned-ade-512-512 on the NICOPOI-9/morphpad_coord_hgo_512_4class_v2 dataset. It achieves the following results on the evaluation set:

Loss: 0.7748
Mean Iou: 0.5948
Mean Accuracy: 0.7423
Overall Accuracy: 0.7459
Accuracy 0-0: 0.6960
Accuracy 0-90: 0.8461
Accuracy 90-0: 0.7033
Accuracy 90-90: 0.7239
Iou 0-0: 0.5939
Iou 0-90: 0.6000
Iou 90-0: 0.5825
Iou 90-90: 0.6029

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: 1
eval_batch_size: 1
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: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy 0-0	Accuracy 0-90	Accuracy 90-0	Accuracy 90-90	Iou 0-0	Iou 0-90	Iou 90-0	Iou 90-90
1.3484	1.3638	4000	1.3289	0.2017	0.3465	0.3565	0.3694	0.4129	0.5017	0.1023	0.1982	0.2542	0.2676	0.0870
0.9532	2.7276	8000	1.2774	0.2355	0.3830	0.3915	0.2214	0.5572	0.3663	0.3871	0.1689	0.2890	0.2712	0.2128
1.2259	4.0914	12000	1.2069	0.2709	0.4320	0.4491	0.2263	0.5930	0.6262	0.2824	0.1906	0.3401	0.3464	0.2064
0.901	5.4552	16000	1.2162	0.2917	0.4547	0.4635	0.2993	0.4156	0.6894	0.4144	0.2374	0.3064	0.3670	0.2558
1.1421	6.8190	20000	1.1904	0.3075	0.4789	0.4844	0.2503	0.3943	0.6817	0.5895	0.2227	0.3170	0.3803	0.3099
0.5421	8.1827	24000	1.1073	0.3455	0.5242	0.5139	0.5857	0.3705	0.4666	0.6741	0.3666	0.3126	0.3624	0.3406
1.2263	9.5465	28000	1.0779	0.3876	0.5563	0.5729	0.3820	0.7883	0.6612	0.3935	0.3299	0.4348	0.4510	0.3348
1.0113	10.9103	32000	0.9747	0.4281	0.5963	0.6055	0.4518	0.6357	0.7514	0.5463	0.3777	0.4580	0.4585	0.4182
0.8357	12.2741	36000	0.9503	0.4550	0.6259	0.6268	0.6729	0.6283	0.6470	0.5555	0.4528	0.4661	0.4677	0.4336
0.8507	13.6379	40000	0.9036	0.4838	0.6491	0.6540	0.6036	0.7493	0.6441	0.5993	0.4728	0.5037	0.4880	0.4708
1.0458	15.0017	44000	0.8932	0.4775	0.6410	0.6532	0.4857	0.7667	0.7653	0.5462	0.4278	0.5066	0.5142	0.4614
0.6563	16.3655	48000	0.8989	0.4839	0.6466	0.6588	0.5395	0.7874	0.7578	0.5016	0.4608	0.5116	0.5228	0.4403
0.6149	17.7293	52000	0.9217	0.4875	0.6542	0.6623	0.4427	0.7185	0.7490	0.7066	0.4078	0.5194	0.5161	0.5067
0.6304	19.0931	56000	0.8176	0.5281	0.6870	0.6938	0.5785	0.7612	0.7479	0.6606	0.5003	0.5449	0.5454	0.5218
0.6309	20.4569	60000	0.8703	0.5127	0.6723	0.6838	0.4812	0.7879	0.7896	0.6305	0.4419	0.5425	0.5378	0.5285
2.9395	21.8207	64000	0.8044	0.5372	0.6935	0.7025	0.5420	0.7885	0.7784	0.6652	0.4866	0.5640	0.5473	0.5509
0.3831	23.1845	68000	0.8331	0.5362	0.6925	0.7013	0.6518	0.7583	0.8161	0.5439	0.5442	0.5586	0.5490	0.4932
0.7022	24.5482	72000	0.7709	0.5627	0.7173	0.7204	0.6734	0.7232	0.7788	0.6937	0.5589	0.5685	0.5622	0.5613
0.7517	25.9120	76000	0.8931	0.5084	0.6688	0.6848	0.4156	0.8377	0.8244	0.5974	0.3966	0.5571	0.5478	0.5323
0.7511	27.2758	80000	0.8619	0.5340	0.6915	0.7023	0.4844	0.7957	0.8032	0.6828	0.4533	0.5674	0.5563	0.5591
0.2229	28.6396	84000	0.9052	0.5252	0.6901	0.6915	0.5348	0.7514	0.6383	0.8359	0.4907	0.5594	0.5282	0.5224
0.457	30.0034	88000	0.7445	0.5770	0.7277	0.7324	0.6864	0.8048	0.7465	0.6730	0.5769	0.5859	0.5785	0.5667
0.1075	31.3672	92000	0.9722	0.5032	0.6636	0.6812	0.4195	0.8790	0.8009	0.5553	0.4012	0.5447	0.5593	0.5076
0.3649	32.7310	96000	0.8135	0.5576	0.7119	0.7204	0.5188	0.7787	0.8167	0.7333	0.4837	0.5869	0.5697	0.5901
0.4757	34.0948	100000	0.8154	0.5715	0.7199	0.7297	0.6141	0.7872	0.8601	0.6181	0.5540	0.6028	0.5726	0.5567
0.5791	35.4586	104000	0.7704	0.5925	0.7397	0.7450	0.7285	0.8175	0.7712	0.6418	0.6013	0.5999	0.5956	0.5732
0.468	36.8224	108000	0.7779	0.5908	0.7406	0.7429	0.6884	0.8257	0.6922	0.7559	0.5883	0.5982	0.5748	0.6020
0.5182	38.1862	112000	0.8067	0.5719	0.7267	0.7308	0.5660	0.7542	0.7750	0.8117	0.5224	0.5922	0.5855	0.5875
0.3732	39.5499	116000	0.7748	0.5948	0.7423	0.7459	0.6960	0.8461	0.7033	0.7239	0.5939	0.6000	0.5825	0.6029

Framework versions

Transformers 4.48.3
Pytorch 2.1.0
Datasets 3.2.0
Tokenizers 0.21.0

NICOPOI-9/segformer-b0-finetuned-ade20k-morphpadver1-hgo-coord-v9_mix_resample_40epochs

segformer-b0-finetuned-ade20k-morphpadver1-hgo-coord-v9_mix_resample_40epochs

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NICOPOI-9/
segformer-b0-finetuned-ade20k-morphpadver1-hgo-coord-v9_mix_resample_40epochs