CoolFace
Modelpublic

mmillet/rubert-tiny2_finetuned_emotion_experiment_augmented_anger_fear_no_emojis

sourceHugging Facemitupdated 4y agoView on Hugging Face
0likes77downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

rubert-tiny2finetunedemotionexperimentaugmentedangerfearnoemojis

This model is a fine-tuned version of cointegrated/rubert-tiny2 on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.5820
  • —Accuracy: 0.7881
  • —F1: 0.7886
  • —Precision: 0.7906
  • —Recall: 0.7881

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: 1e-05
  • —trainbatchsize: 64
  • —evalbatchsize: 64
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 40

Training results

Training LossEpochStepValidation LossAccuracyF1PrecisionRecall
1.09961.0691.00130.68790.67790.70700.6879
0.95242.01380.86510.72650.72450.73220.7265
0.83453.02070.78210.74220.74130.74450.7422
0.75734.02760.72220.74840.74730.74820.7484
0.69235.03450.68280.75680.75620.75620.7568
0.64126.04140.65310.75680.75590.75560.7568
0.59827.04830.63200.76100.76010.75970.7610
0.55938.05520.61330.76510.76550.76640.7651
0.51839.06210.60360.77140.77080.77090.7714
0.504210.06900.59510.77560.77550.77600.7756
0.48311.07590.58780.77660.77680.77740.7766
0.453112.08280.58550.78500.78410.78390.7850
0.438613.08970.58280.77970.77900.77860.7797
0.423814.09660.57880.77770.77800.77860.7777
0.401815.010350.57930.78390.78420.78550.7839
0.399816.011040.58010.78500.78440.78410.7850
0.374717.011730.57910.78390.78360.78330.7839
0.359518.012420.57990.78910.78910.78940.7891
0.357519.013110.58200.78810.78860.79060.7881

Framework versions

  • —Transformers 4.19.2
  • —Pytorch 1.11.0+cu113
  • —Datasets 2.2.2
  • —Tokenizers 0.12.1