CoolFace
Modelpublic

gokuls/hBERTv2_new_no_pretrain_rte

sourceHugging Faceupdated 3y 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. -->

hBERTv2newnopretrainrte

This model is a fine-tuned version of [](https://huggingface.co/) on the GLUE RTE dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.6981
  • —Accuracy: 0.5307

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
0.76971.0200.75260.5271
0.72852.0400.72080.5271
0.72013.0600.71120.5343
0.70434.0800.69810.5307
0.65695.01000.72510.5235
0.57626.01200.85710.4765
0.43367.01400.95400.4765
0.32998.01601.24640.4838
0.25619.01801.42990.5018

Framework versions

  • —Transformers 4.30.2
  • —Pytorch 1.14.0a0+410ce96
  • —Datasets 2.12.0
  • —Tokenizers 0.13.3