spellcorrector_20_02_050_qwerty_v14

This model is a fine-tuned version of Buseak/spellcorrector_20_02_050_qwerty_v11 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0000
• Precision: 1.0
• Recall: 1.0
• F1: 1.0
• 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0331	1.0	967	0.0132	0.9919	0.9893	0.9906	0.9963
0.019	2.0	1934	0.0077	0.9941	0.9936	0.9938	0.9977
0.0127	3.0	2901	0.0060	0.9973	0.9973	0.9973	0.9982
0.0102	4.0	3868	0.0035	0.9968	0.9962	0.9965	0.9990
0.0077	5.0	4835	0.0024	0.9989	0.9989	0.9989	0.9993
0.0065	6.0	5802	0.0015	0.9989	0.9984	0.9987	0.9996
0.0047	7.0	6769	0.0012	0.9995	0.9989	0.9992	0.9997
0.0036	8.0	7736	0.0009	1.0	0.9995	0.9997	0.9997
0.0031	9.0	8703	0.0007	0.9995	0.9995	0.9995	0.9998
0.0027	10.0	9670	0.0004	1.0	1.0	1.0	0.9999
0.0021	11.0	10637	0.0003	1.0	1.0	1.0	0.9999
0.0017	12.0	11604	0.0002	1.0	1.0	1.0	1.0000
0.0015	13.0	12571	0.0001	1.0	1.0	1.0	1.0000
0.0011	14.0	13538	0.0001	1.0	1.0	1.0	1.0000
0.001	15.0	14505	0.0001	1.0	1.0	1.0	1.0000
0.0008	16.0	15472	0.0001	1.0	1.0	1.0	1.0000
0.0007	17.0	16439	0.0000	1.0	1.0	1.0	1.0000
0.0006	18.0	17406	0.0000	1.0	1.0	1.0	1.0000
0.0006	19.0	18373	0.0000	1.0	1.0	1.0	1.0000
0.0005	20.0	19340	0.0000	1.0	1.0	1.0	1.0000

Framework versions

• Transformers 4.37.2
• Pytorch 2.1.0+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2