Edit model card

spellcorrector_11_02_050_1_per_word_v5

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: 0.0399
  • Precision: 0.9989
  • Recall: 0.9946
  • F1: 0.9968
  • Accuracy: 0.9880

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: 5e-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
0.3884 1.0 967 0.1563 0.9714 0.9635 0.9674 0.9611
0.1648 2.0 1934 0.1297 0.9784 0.9716 0.9750 0.9669
0.1431 3.0 2901 0.1157 0.9924 0.9753 0.9838 0.9698
0.1286 4.0 3868 0.1042 0.9897 0.9807 0.9852 0.9722
0.1201 5.0 4835 0.0969 0.9903 0.9839 0.9871 0.9737
0.1134 6.0 5802 0.0882 0.9903 0.9861 0.9882 0.9757
0.106 7.0 6769 0.0808 0.9935 0.9855 0.9895 0.9773
0.1002 8.0 7736 0.0763 0.9924 0.9861 0.9892 0.9786
0.0945 9.0 8703 0.0696 0.9957 0.9855 0.9906 0.9799
0.0903 10.0 9670 0.0641 0.9919 0.9893 0.9906 0.9813
0.0866 11.0 10637 0.0597 0.9920 0.9925 0.9922 0.9825
0.0822 12.0 11604 0.0557 0.9962 0.9925 0.9944 0.9835
0.0787 13.0 12571 0.0523 0.9978 0.9914 0.9946 0.9843
0.0751 14.0 13538 0.0500 0.9984 0.9946 0.9965 0.9852
0.0715 15.0 14505 0.0467 0.9968 0.9946 0.9957 0.9861
0.0698 16.0 15472 0.0438 0.9995 0.9952 0.9973 0.9868
0.0674 17.0 16439 0.0426 0.9984 0.9952 0.9968 0.9870
0.0652 18.0 17406 0.0410 0.9989 0.9952 0.9970 0.9875
0.0639 19.0 18373 0.0403 0.9989 0.9946 0.9968 0.9879
0.0628 20.0 19340 0.0399 0.9989 0.9946 0.9968 0.9880

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.17.0
  • Tokenizers 0.15.1
Downloads last month
6
Safetensors
Model size
132M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for Buseak/spellcorrector_11_02_050_1_per_word_v5

Base model

google/canine-s
Finetuned
(8)
this model
Finetunes
1 model