README.md

metadata

library_name: transformers
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
model-index:
  - name: UNetOscillatoryNeuron
    results: []

UNetOscillatoryNeuron

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

• Loss: 0.0222
• Dice: 0.6912
• Iou: 0.5379
• Precision: 0.9949
• Recall: 0.5395

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: 0.001
• train_batch_size: 8
• eval_batch_size: 8
• 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
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Dice	Iou	Precision	Recall
0.9974	1.0	27	6.4241	0.4013	0.2537	0.2590	0.9319
0.4597	2.0	54	0.5612	0.6393	0.4714	0.5705	0.7437
0.1191	3.0	81	0.1381	0.7612	0.6225	0.8853	0.6797
0.064	4.0	108	0.1400	0.7652	0.6276	0.8892	0.6825
0.0473	5.0	135	0.0575	0.7339	0.5885	0.9563	0.6051
0.0438	6.0	162	0.0864	0.7435	0.5995	0.9429	0.6212
0.0366	7.0	189	0.0374	0.6901	0.5353	0.9899	0.5384
0.0345	8.0	216	0.0530	0.7390	0.5943	0.9591	0.6092
0.0326	9.0	243	0.0351	0.7177	0.5690	0.9799	0.5761
0.0311	10.0	270	0.0345	0.7204	0.5714	0.9780	0.5787
0.028	11.0	297	0.0582	0.7436	0.6005	0.9527	0.6192
0.0272	12.0	324	0.0304	0.7126	0.5623	0.9838	0.5675
0.0269	13.0	351	0.0291	0.7126	0.5626	0.9832	0.5682
0.0269	14.0	378	0.0273	0.7053	0.5537	0.9886	0.5574
0.0264	15.0	405	0.0290	0.7202	0.5717	0.9837	0.5772
0.0238	16.0	432	0.0359	0.7309	0.5844	0.9743	0.5931
0.0247	17.0	459	0.0272	0.7204	0.5718	0.9844	0.5770
0.024	18.0	486	0.0419	0.7354	0.5902	0.9665	0.6029
0.021	19.0	513	0.0457	0.7404	0.5963	0.9627	0.6103
0.0217	20.0	540	0.0257	0.7176	0.5683	0.9855	0.5729
0.0228	21.0	567	0.0293	0.7195	0.5708	0.9801	0.5777
0.0218	22.0	594	0.0309	0.7271	0.5801	0.9782	0.5878
0.0207	23.0	621	0.0305	0.7267	0.5797	0.9783	0.5873
0.0192	24.0	648	0.0234	0.7141	0.5645	0.9881	0.5685
0.019	25.0	675	0.0247	0.7192	0.5706	0.9860	0.5752
0.0194	26.0	702	0.0251	0.7183	0.5695	0.9855	0.5744
0.0194	27.0	729	0.0220	0.7051	0.5535	0.9914	0.5563
0.0203	28.0	756	0.0237	0.7130	0.5635	0.9876	0.5678
0.019	29.0	783	0.0230	0.7157	0.5666	0.9881	0.5703
0.0176	30.0	810	0.0241	0.7198	0.5711	0.9865	0.5754
0.0168	31.0	837	0.0261	0.7235	0.5759	0.9835	0.5815
0.0169	32.0	864	0.0264	0.7213	0.5732	0.9826	0.5794
0.0165	33.0	891	0.0243	0.7196	0.5710	0.9849	0.5760
0.0163	34.0	918	0.0213	0.7033	0.5514	0.9915	0.5541
0.0161	35.0	945	0.0214	0.7057	0.5544	0.9906	0.5574
0.0153	36.0	972	0.0208	0.7093	0.5587	0.9908	0.5616
0.0152	37.0	999	0.0218	0.7101	0.5595	0.9898	0.5627
0.0151	38.0	1026	0.0224	0.7160	0.5666	0.9875	0.5707
0.0145	39.0	1053	0.0204	0.7015	0.5497	0.9927	0.5521
0.0143	40.0	1080	0.0208	0.7035	0.5519	0.9921	0.5545
0.014	41.0	1107	0.0205	0.7015	0.5496	0.9932	0.5517
0.0142	42.0	1134	0.0207	0.7027	0.5512	0.9924	0.5536
0.014	43.0	1161	0.0214	0.7121	0.5623	0.9892	0.5658
0.0136	44.0	1188	0.0213	0.6950	0.5423	0.9944	0.5441
0.0134	45.0	1215	0.0212	0.6980	0.5456	0.9935	0.5477
0.0133	46.0	1242	0.0214	0.6946	0.5420	0.9943	0.5438
0.013	47.0	1269	0.0210	0.6959	0.5432	0.9945	0.5449
0.013	48.0	1296	0.0221	0.6919	0.5387	0.9949	0.5403
0.0129	49.0	1323	0.0219	0.6925	0.5393	0.9945	0.5410
0.0125	50.0	1350	0.0222	0.6912	0.5379	0.9949	0.5395

Framework versions

• Transformers 4.51.1
• Pytorch 2.5.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.0