README.md

metadata

license: other
tags:
  - generated_from_trainer
model-index:
  - name: segformer-b0-finetuned-segments-toolwear
    results: []

segformer-b0-finetuned-segments-toolwear

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

• Loss: 0.0799
• Mean Iou: 0.4629
• Mean Accuracy: 0.9258
• Overall Accuracy: 0.9258
• Accuracy Unlabeled: nan
• Accuracy Liver: 0.9258
• Iou Unlabeled: 0.0
• Iou Liver: 0.9258

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: 0.0001
• train_batch_size: 24
• eval_batch_size: 24
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 35

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Liver	Iou Unlabeled	Iou Liver
0.2837	0.8	20	0.3699	0.3876	0.7752	0.7752	nan	0.7752	0.0	0.7752
0.2264	1.6	40	0.1982	0.4222	0.8444	0.8444	nan	0.8444	0.0	0.8444
0.1687	2.4	60	0.1594	0.3988	0.7977	0.7977	nan	0.7977	0.0	0.7977
0.1489	3.2	80	0.1396	0.4050	0.8100	0.8100	nan	0.8100	0.0	0.8100
0.1111	4.0	100	0.1203	0.4223	0.8446	0.8446	nan	0.8446	0.0	0.8446
0.1115	4.8	120	0.1160	0.4512	0.9023	0.9023	nan	0.9023	0.0	0.9023
0.1081	5.6	140	0.1053	0.4504	0.9009	0.9009	nan	0.9009	0.0	0.9009
0.1111	6.4	160	0.0960	0.4526	0.9051	0.9051	nan	0.9051	0.0	0.9051
0.0904	7.2	180	0.0954	0.4646	0.9292	0.9292	nan	0.9292	0.0	0.9292
0.0868	8.0	200	0.0925	0.4593	0.9187	0.9187	nan	0.9187	0.0	0.9187
0.092	8.8	220	0.0852	0.4630	0.9261	0.9261	nan	0.9261	0.0	0.9261
0.0686	9.6	240	0.0897	0.4631	0.9263	0.9263	nan	0.9263	0.0	0.9263
0.0684	10.4	260	0.0939	0.4727	0.9455	0.9455	nan	0.9455	0.0	0.9455
0.0634	11.2	280	0.0919	0.4241	0.8483	0.8483	nan	0.8483	0.0	0.8483
0.059	12.0	300	0.0886	0.4727	0.9455	0.9455	nan	0.9455	0.0	0.9455
0.052	12.8	320	0.0764	0.4554	0.9108	0.9108	nan	0.9108	0.0	0.9108
0.0558	13.6	340	0.0769	0.4629	0.9258	0.9258	nan	0.9258	0.0	0.9258
0.0594	14.4	360	0.0770	0.4616	0.9231	0.9231	nan	0.9231	0.0	0.9231
0.0641	15.2	380	0.0844	0.4709	0.9417	0.9417	nan	0.9417	0.0	0.9417
0.0645	16.0	400	0.0790	0.4632	0.9263	0.9263	nan	0.9263	0.0	0.9263
0.0545	16.8	420	0.0776	0.4610	0.9220	0.9220	nan	0.9220	0.0	0.9220
0.056	17.6	440	0.0780	0.4541	0.9082	0.9082	nan	0.9082	0.0	0.9082
0.0472	18.4	460	0.0742	0.4595	0.9189	0.9189	nan	0.9189	0.0	0.9189
0.0478	19.2	480	0.0806	0.4690	0.9380	0.9380	nan	0.9380	0.0	0.9380
0.0523	20.0	500	0.0741	0.4550	0.9100	0.9100	nan	0.9100	0.0	0.9100
0.0401	20.8	520	0.0794	0.4637	0.9274	0.9274	nan	0.9274	0.0	0.9274
0.041	21.6	540	0.0772	0.4631	0.9262	0.9262	nan	0.9262	0.0	0.9262
0.0386	22.4	560	0.0795	0.4620	0.9240	0.9240	nan	0.9240	0.0	0.9240
0.0386	23.2	580	0.0761	0.4616	0.9232	0.9232	nan	0.9232	0.0	0.9232
0.0628	24.0	600	0.0778	0.4636	0.9271	0.9271	nan	0.9271	0.0	0.9271
0.0387	24.8	620	0.0782	0.4613	0.9226	0.9226	nan	0.9226	0.0	0.9226
0.0422	25.6	640	0.0778	0.4616	0.9233	0.9233	nan	0.9233	0.0	0.9233
0.0392	26.4	660	0.0838	0.4696	0.9393	0.9393	nan	0.9393	0.0	0.9393
0.04	27.2	680	0.0809	0.4658	0.9315	0.9315	nan	0.9315	0.0	0.9315
0.0341	28.0	700	0.0822	0.4667	0.9335	0.9335	nan	0.9335	0.0	0.9335
0.0329	28.8	720	0.0797	0.4639	0.9278	0.9278	nan	0.9278	0.0	0.9278
0.0373	29.6	740	0.0769	0.4582	0.9163	0.9163	nan	0.9163	0.0	0.9163
0.0366	30.4	760	0.0804	0.4632	0.9264	0.9264	nan	0.9264	0.0	0.9264
0.0432	31.2	780	0.0793	0.4587	0.9174	0.9174	nan	0.9174	0.0	0.9174
0.0328	32.0	800	0.0838	0.4688	0.9377	0.9377	nan	0.9377	0.0	0.9377
0.0377	32.8	820	0.0805	0.4643	0.9286	0.9286	nan	0.9286	0.0	0.9286
0.0327	33.6	840	0.0784	0.4614	0.9228	0.9228	nan	0.9228	0.0	0.9228
0.032	34.4	860	0.0799	0.4629	0.9258	0.9258	nan	0.9258	0.0	0.9258

Framework versions

• Transformers 4.28.0
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.13.3