CoolFace
Modelpublic

gokuls/hBERTv1_new_pretrain_sst2

sourceHugging Faceupdated 3y agoView on Hugging Face
0likes17downloads
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. -->

hBERTv1newpretrain_sst2

This model is a fine-tuned version of gokuls/bert_12_layer_model_v1_complete_training_new on the GLUE SST2 dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.4752
  • —Accuracy: 0.7878

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: 4e-05
  • —trainbatchsize: 128
  • —evalbatchsize: 128
  • —seed: 10
  • —distributed_type: multi-GPU
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 50

Training results

Training LossEpochStepValidation LossAccuracy
0.42581.05270.49940.8062
0.26522.010540.56330.8005
0.22143.015810.47520.7878
0.20144.021080.53290.7890
0.18135.026350.54100.7924
0.16796.031620.58570.8085
0.15267.036890.76540.8039
0.14058.042160.67150.7878

Framework versions

  • —Transformers 4.29.2
  • —Pytorch 1.14.0a0+410ce96
  • —Datasets 2.12.0
  • —Tokenizers 0.13.3