CoolFace
Modelpublic

gokuls/distilbert_add_GLUE_Experiment_logit_kd_cola

sourceHugging Faceapache-2.0updated 4y agoView on Hugging Face
0likes23downloads
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. -->

distilbertaddGLUEExperimentlogitkdcola

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

  • —Loss: 0.6820
  • —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: 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 LossMatthews Correlation
0.82131.0340.68750.0
0.7992.0680.68400.0
0.79713.01020.68910.0
0.79774.01360.68400.0
0.80075.01700.68620.0
0.79876.02040.68440.0
0.79897.02380.68710.0
0.79568.02720.68200.0
0.77289.03060.73890.0
0.71110.03400.75920.0493
0.686611.03740.73080.0704
0.667112.04080.73930.0670
0.643913.04420.78020.0595

Framework versions

  • —Transformers 4.26.0
  • —Pytorch 1.14.0a0+410ce96
  • —Datasets 2.9.0
  • —Tokenizers 0.13.2