Edit model card

hbertv2-emotion_48_w_in

This model is a fine-tuned version of gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48 on the emotion dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1958
  • Accuracy: 0.931

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: 64
  • eval_batch_size: 64
  • seed: 33
  • distributed_type: multi-GPU
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.6416 1.0 250 0.2852 0.8945
0.2565 2.0 500 0.1956 0.9205
0.1769 3.0 750 0.2171 0.9225
0.1447 4.0 1000 0.1815 0.9225
0.1192 5.0 1250 0.1960 0.9235
0.1023 6.0 1500 0.1958 0.931
0.0811 7.0 1750 0.2533 0.928
0.0698 8.0 2000 0.3094 0.9305
0.0547 9.0 2250 0.2719 0.93
0.0442 10.0 2500 0.3100 0.926

Framework versions

  • Transformers 4.30.2
  • Pytorch 1.14.0a0+410ce96
  • Datasets 2.13.0
  • Tokenizers 0.13.3
Downloads last month
7

Dataset used to train gokuls/hbertv2-emotion_48_w_in

Evaluation results