sentence_splitter_final_v2

This model is a fine-tuned version of google/canine-s on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0000
• Precision: 0.8
• Recall: 1.0
• F1: 0.8889
• Accuracy: 1.0000

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	95	0.0037	0.0690	0.5	0.1212	0.9988
No log	2.0	190	0.0022	0.0909	1.0	0.1667	0.9993
No log	3.0	285	0.0014	0.1333	1.0	0.2353	0.9995
No log	4.0	380	0.0010	0.1905	1.0	0.32	0.9996
No log	5.0	475	0.0008	0.25	1.0	0.4	0.9997
0.0096	6.0	570	0.0004	0.3636	1.0	0.5333	0.9998
0.0096	7.0	665	0.0004	0.2222	1.0	0.3636	0.9999
0.0096	8.0	760	0.0002	0.4	1.0	0.5714	0.9999
0.0096	9.0	855	0.0003	0.1905	1.0	0.32	0.9999
0.0096	10.0	950	0.0003	0.2105	1.0	0.3478	0.9999
0.0008	11.0	1045	0.0001	0.3333	1.0	0.5	1.0000
0.0008	12.0	1140	0.0001	0.5	1.0	0.6667	1.0000
0.0008	13.0	1235	0.0001	0.4444	1.0	0.6154	1.0000
0.0008	14.0	1330	0.0000	0.8	1.0	0.8889	1.0000
0.0008	15.0	1425	0.0000	0.6667	1.0	0.8	1.0000
0.0003	16.0	1520	0.0000	0.8	1.0	0.8889	1.0000
0.0003	17.0	1615	0.0000	0.8	1.0	0.8889	1.0000
0.0003	18.0	1710	0.0000	0.8	1.0	0.8889	1.0000
0.0003	19.0	1805	0.0000	0.8	1.0	0.8889	1.0000
0.0003	20.0	1900	0.0000	0.8	1.0	0.8889	1.0000
0.0003	21.0	1995	0.0000	0.8	1.0	0.8889	1.0000
0.0001	22.0	2090	0.0000	0.8	1.0	0.8889	1.0000
0.0001	23.0	2185	0.0000	0.8	1.0	0.8889	1.0000
0.0001	24.0	2280	0.0000	0.8	1.0	0.8889	1.0000
0.0001	25.0	2375	0.0000	0.8	1.0	0.8889	1.0000

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0