CoolFace
Modelpublic

gokuls/HBERTv1_48_L8_H512_A8_massive

sourceHugging Faceupdated 3y agoView on Hugging Face
0likes7downloads
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. -->

HBERTv148L8H512A8_massive

This model is a fine-tuned version of gokuls/HBERTv1_48_L8_H512_A8 on the massive dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.8009
  • —Accuracy: 0.8569

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: 64
  • —evalbatchsize: 64
  • —seed: 33
  • —distributed_type: multi-GPU
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 15

Training results

Training LossEpochStepValidation LossAccuracy
3.23831.01802.01100.4747
1.50512.03601.09040.7093
0.89983.05400.85440.7727
0.6614.07200.70290.8160
0.50525.09000.69870.8131
0.38896.010800.69010.8244
0.30627.012600.67460.8352
0.24228.014400.69460.8396
0.18349.016200.72900.8441
0.140210.018000.74160.8406
0.109811.019800.78280.8387
0.080312.021600.77000.8460
0.061213.023400.78910.8515
0.046814.025200.80090.8569
0.037515.027000.80320.8569

Framework versions

  • —Transformers 4.34.0
  • —Pytorch 1.14.0a0+410ce96
  • —Datasets 2.14.5
  • —Tokenizers 0.14.0