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: 0.0223
• Mean Iou: 0.4979
• Mean Accuracy: 0.9957
• Overall Accuracy: 0.9957
• Accuracy Unlabeled: nan
• Accuracy Tool: 0.9957
• Iou Unlabeled: 0.0
• Iou Tool: 0.9957

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: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Tool	Iou Unlabeled	Iou Tool
0.1534	1.18	20	0.3425	0.4977	0.9955	0.9955	nan	0.9955	0.0	0.9955
0.091	2.35	40	0.1076	0.4948	0.9897	0.9897	nan	0.9897	0.0	0.9897
0.0827	3.53	60	0.0828	0.4965	0.9931	0.9931	nan	0.9931	0.0	0.9931
0.0729	4.71	80	0.0795	0.4967	0.9934	0.9934	nan	0.9934	0.0	0.9934
0.0825	5.88	100	0.0606	0.4910	0.9819	0.9819	nan	0.9819	0.0	0.9819
0.0604	7.06	120	0.0546	0.4910	0.9820	0.9820	nan	0.9820	0.0	0.9820
0.0575	8.24	140	0.0460	0.4942	0.9884	0.9884	nan	0.9884	0.0	0.9884
0.0592	9.41	160	0.0450	0.4906	0.9813	0.9813	nan	0.9813	0.0	0.9813
0.0478	10.59	180	0.0400	0.4981	0.9962	0.9962	nan	0.9962	0.0	0.9962
0.046	11.76	200	0.0403	0.4982	0.9964	0.9964	nan	0.9964	0.0	0.9964
0.0535	12.94	220	0.0340	0.4971	0.9941	0.9941	nan	0.9941	0.0	0.9941
0.0317	14.12	240	0.0332	0.4975	0.9949	0.9949	nan	0.9949	0.0	0.9949
0.0352	15.29	260	0.0328	0.4982	0.9964	0.9964	nan	0.9964	0.0	0.9964
0.0258	16.47	280	0.0295	0.4963	0.9926	0.9926	nan	0.9926	0.0	0.9926
0.0218	17.65	300	0.0265	0.4968	0.9935	0.9935	nan	0.9935	0.0	0.9935
0.026	18.82	320	0.0284	0.4979	0.9958	0.9958	nan	0.9958	0.0	0.9958
0.026	20.0	340	0.0267	0.4971	0.9941	0.9941	nan	0.9941	0.0	0.9941
0.02	21.18	360	0.0242	0.4967	0.9935	0.9935	nan	0.9935	0.0	0.9935
0.0255	22.35	380	0.0270	0.4975	0.9949	0.9949	nan	0.9949	0.0	0.9949
0.0282	23.53	400	0.0240	0.4973	0.9946	0.9946	nan	0.9946	0.0	0.9946
0.0188	24.71	420	0.0244	0.4972	0.9944	0.9944	nan	0.9944	0.0	0.9944
0.0196	25.88	440	0.0226	0.4961	0.9922	0.9922	nan	0.9922	0.0	0.9922
0.0165	27.06	460	0.0235	0.4968	0.9937	0.9937	nan	0.9937	0.0	0.9937
0.02	28.24	480	0.0245	0.4981	0.9962	0.9962	nan	0.9962	0.0	0.9962
0.0213	29.41	500	0.0225	0.4972	0.9944	0.9944	nan	0.9944	0.0	0.9944
0.0174	30.59	520	0.0221	0.4970	0.9940	0.9940	nan	0.9940	0.0	0.9940
0.0163	31.76	540	0.0226	0.4975	0.9951	0.9951	nan	0.9951	0.0	0.9951
0.0242	32.94	560	0.0236	0.4978	0.9956	0.9956	nan	0.9956	0.0	0.9956
0.0195	34.12	580	0.0217	0.4976	0.9953	0.9953	nan	0.9953	0.0	0.9953
0.0134	35.29	600	0.0220	0.4974	0.9948	0.9948	nan	0.9948	0.0	0.9948
0.0192	36.47	620	0.0216	0.4974	0.9947	0.9947	nan	0.9947	0.0	0.9947
0.0138	37.65	640	0.0219	0.4974	0.9948	0.9948	nan	0.9948	0.0	0.9948
0.0147	38.82	660	0.0215	0.4973	0.9945	0.9945	nan	0.9945	0.0	0.9945
0.0208	40.0	680	0.0219	0.4979	0.9958	0.9958	nan	0.9958	0.0	0.9958
0.0152	41.18	700	0.0211	0.4974	0.9948	0.9948	nan	0.9948	0.0	0.9948
0.0145	42.35	720	0.0214	0.4977	0.9954	0.9954	nan	0.9954	0.0	0.9954
0.0138	43.53	740	0.0217	0.4977	0.9954	0.9954	nan	0.9954	0.0	0.9954
0.0122	44.71	760	0.0218	0.4977	0.9954	0.9954	nan	0.9954	0.0	0.9954
0.0201	45.88	780	0.0220	0.4976	0.9953	0.9953	nan	0.9953	0.0	0.9953
0.0147	47.06	800	0.0219	0.4977	0.9954	0.9954	nan	0.9954	0.0	0.9954
0.0131	48.24	820	0.0213	0.4975	0.9950	0.9950	nan	0.9950	0.0	0.9950
0.016	49.41	840	0.0223	0.4979	0.9957	0.9957	nan	0.9957	0.0	0.9957

Framework versions

• Transformers 4.28.0
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.13.3