segformer-b5-finetuned-morphpad-morphpadver1-hgo-coord-v9_16mat_2void_40epochs

This model is a fine-tuned version of NICOPOI-9/segformer-b5-finetuned-ade20k-morphpadver1-hgo-coord-v9_mix_resample_40epochs on the NICOPOI-9/morphpad_16_labels_2voids_gaussian dataset. It achieves the following results on the evaluation set:

• Loss: 0.3575
• Mean Iou: 0.8930
• Mean Accuracy: 0.9416
• Overall Accuracy: 0.9463
• Accuracy [0,0]: 0.9359
• Accuracy [0,1]: 0.9388
• Accuracy [1,0]: 0.9489
• Accuracy [1,1]: 0.9515
• Accuracy [0,2]: 0.9458
• Accuracy [0,3]: 0.9591
• Accuracy [1,2]: 0.9443
• Accuracy [1,3]: 0.9617
• Accuracy [2,0]: 0.9116
• Accuracy [2,1]: 0.9338
• Accuracy [2,2]: 0.9560
• Accuracy [2,3]: 0.9303
• Accuracy [3,0]: 0.9475
• Accuracy [3,1]: 0.9086
• Accuracy [3,2]: 0.9391
• Accuracy [3,3]: 0.9160
• Accuracy Void: 0.9788
• Iou [0,0]: 0.8957
• Iou [0,1]: 0.9112
• Iou [1,0]: 0.8826
• Iou [1,1]: 0.9078
• Iou [0,2]: 0.8907
• Iou [0,3]: 0.9174
• Iou [1,2]: 0.8766
• Iou [1,3]: 0.8767
• Iou [2,0]: 0.8790
• Iou [2,1]: 0.9008
• Iou [2,2]: 0.8806
• Iou [2,3]: 0.8972
• Iou [3,0]: 0.9088
• Iou [3,1]: 0.8467
• Iou [3,2]: 0.8877
• Iou [3,3]: 0.8755
• Iou Void: 0.9455

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 [0,1]	Accuracy [1,0]	Accuracy [1,1]	Accuracy [0,2]	Accuracy [0,3]	Accuracy [1,2]	Accuracy [1,3]	Accuracy [2,0]	Accuracy [2,1]	Accuracy [2,2]	Accuracy [2,3]	Accuracy [3,0]	Accuracy [3,1]	Accuracy [3,2]	Accuracy [3,3]	Accuracy Void	Iou [0,0]	Iou [0,1]	Iou [1,0]	Iou [1,1]	Iou [0,2]	Iou [0,3]	Iou [1,2]	Iou [1,3]	Iou [2,0]	Iou [2,1]	Iou [2,2]	Iou [2,3]	Iou [3,0]	Iou [3,1]	Iou [3,2]	Iou [3,3]	Iou Void
0.2611	7.3260	4000	0.4402	0.7621	0.8618	0.8738	0.8911	0.8941	0.9373	0.8466	0.8823	0.8020	0.8669	0.8736	0.8264	0.8524	0.8802	0.8070	0.8686	0.7729	0.8628	0.8363	0.9506	0.7294	0.8149	0.7502	0.7804	0.7516	0.7345	0.7654	0.7698	0.7054	0.7860	0.7727	0.7222	0.7650	0.7060	0.7560	0.7201	0.9259
0.0752	14.6520	8000	0.3003	0.8459	0.9150	0.9219	0.9089	0.9053	0.9341	0.9189	0.9172	0.9234	0.8981	0.9248	0.9091	0.8928	0.9286	0.9263	0.9165	0.8866	0.9078	0.8821	0.9748	0.8681	0.8589	0.8396	0.8530	0.8516	0.8442	0.8280	0.8354	0.8609	0.8350	0.8572	0.8455	0.8423	0.7927	0.8292	0.8006	0.9391
0.0345	21.9780	12000	0.3628	0.8687	0.9270	0.9333	0.9331	0.9187	0.9463	0.9345	0.9437	0.9366	0.9222	0.9471	0.9114	0.9126	0.9537	0.9311	0.9123	0.8793	0.9177	0.8809	0.9784	0.8684	0.8907	0.8609	0.8826	0.8744	0.8767	0.8494	0.8459	0.8656	0.8727	0.8758	0.8759	0.8715	0.8379	0.8433	0.8352	0.9405
0.0422	29.3040	16000	0.3604	0.8848	0.9367	0.9422	0.9402	0.9259	0.9539	0.9297	0.9421	0.9427	0.9482	0.9681	0.8912	0.9222	0.9529	0.9482	0.9337	0.8971	0.9419	0.9112	0.9746	0.8744	0.9078	0.8724	0.8885	0.8864	0.8982	0.8794	0.8767	0.8503	0.8826	0.8739	0.9045	0.8934	0.8480	0.8898	0.8710	0.9449
0.0304	36.6300	20000	0.3575	0.8930	0.9416	0.9463	0.9359	0.9388	0.9489	0.9515	0.9458	0.9591	0.9443	0.9617	0.9116	0.9338	0.9560	0.9303	0.9475	0.9086	0.9391	0.9160	0.9788	0.8957	0.9112	0.8826	0.9078	0.8907	0.9174	0.8766	0.8767	0.8790	0.9008	0.8806	0.8972	0.9088	0.8467	0.8877	0.8755	0.9455

Framework versions

• Transformers 4.48.3
• Pytorch 2.1.0
• Datasets 3.2.0
• Tokenizers 0.21.0