CoolFace
Modelpublic

gokuls/mobilebert_sa_GLUE_Experiment_cola_256

sourceHugging Faceapache-2.0updated 4y agoView on Hugging Face
0likes25downloads
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. -->

mobilebertsaGLUEExperimentcola_256

This model is a fine-tuned version of google/mobilebert-uncased on the GLUE COLA dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.6144
  • —Matthews Correlation: 0.0

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: 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 LossMatthews Correlation
0.61351.0670.61780.0
0.60792.01340.61780.0
0.60733.02010.61810.0
0.60664.02680.61670.0
0.60495.03350.61440.0
0.56996.04020.61940.1196
0.50157.04690.67240.1179
0.46688.05360.77230.1198
0.44259.06030.70530.0810
0.427210.06700.83890.1207

Framework versions

  • —Transformers 4.26.0
  • —Pytorch 1.14.0a0+410ce96
  • —Datasets 2.8.0
  • —Tokenizers 0.13.2