CoolFace
Modelpublic

gokuls/hBERTv2_data_aug_mnli

sourceHugging Faceupdated 4y agoView on Hugging Face
0likes15downloads
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. -->

hBERTv2dataaug_mnli

This model is a fine-tuned version of gokuls/bert_12_layer_model_v2 on the GLUE MNLI dataset. It achieves the following results on the evaluation set:

  • —Loss: 1.0988
  • —Accuracy: 0.3182

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
  • —trainbatchsize: 256
  • —evalbatchsize: 256
  • —seed: 10
  • —distributed_type: multi-GPU
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 50
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossAccuracy
1.09881.0314401.09880.3182
1.09852.0628801.09920.3182
1.09853.0943201.09910.3182
1.09854.01257601.09910.3182
1.09855.01572001.09880.3182
1.09856.01886401.09880.3182
1.09857.02200801.09880.3182
1.09858.02515201.09880.3182
1.09859.02829601.09880.3182
1.098510.03144001.09880.3182
1.098511.03458401.09880.3182
1.098512.03772801.09880.3182
1.098513.04087201.09880.3182
1.098514.04401601.09880.3182
1.098515.04716001.09880.3182

Framework versions

  • —Transformers 4.26.1
  • —Pytorch 1.14.0a0+410ce96
  • —Datasets 2.10.1
  • —Tokenizers 0.13.2