Edit model card

vowelizer_1203_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: 0.0018
  • Precision: 0.9989
  • Recall: 0.9988
  • F1: 0.9989
  • Accuracy: 0.9995

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: 4
  • eval_batch_size: 4
  • 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
0.1589 1.0 1933 0.1247 0.9263 0.8773 0.9011 0.9577
0.1227 2.0 3866 0.0937 0.9453 0.9110 0.9278 0.9676
0.1019 3.0 5799 0.0738 0.9589 0.9261 0.9422 0.9743
0.0868 4.0 7732 0.0595 0.9654 0.9530 0.9592 0.9792
0.0745 5.0 9665 0.0470 0.9741 0.9609 0.9675 0.9833
0.0638 6.0 11598 0.0364 0.9799 0.9728 0.9764 0.9873
0.0529 7.0 13531 0.0282 0.9853 0.9748 0.9800 0.9899
0.0473 8.0 15464 0.0218 0.9894 0.9838 0.9866 0.9923
0.0381 9.0 17397 0.0170 0.9909 0.9895 0.9902 0.9940
0.0325 10.0 19330 0.0128 0.9936 0.9921 0.9928 0.9956
0.0284 11.0 21263 0.0100 0.9950 0.9938 0.9944 0.9965
0.0256 12.0 23196 0.0079 0.9959 0.9949 0.9954 0.9972
0.0222 13.0 25129 0.0058 0.9969 0.9965 0.9967 0.9980
0.0196 14.0 27062 0.0048 0.9974 0.9973 0.9974 0.9984
0.016 15.0 28995 0.0036 0.9979 0.9974 0.9977 0.9988
0.0143 16.0 30928 0.0030 0.9983 0.9981 0.9982 0.9990
0.0134 17.0 32861 0.0025 0.9986 0.9984 0.9985 0.9992
0.0117 18.0 34794 0.0021 0.9987 0.9986 0.9987 0.9993
0.0102 19.0 36727 0.0019 0.9987 0.9987 0.9987 0.9994
0.0098 20.0 38660 0.0018 0.9989 0.9988 0.9989 0.9995

Framework versions

  • Transformers 4.28.0
  • Pytorch 2.1.0+cu121
  • Datasets 2.18.0
  • Tokenizers 0.13.3
Downloads last month
4
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.