CoolFace
Modelpublic

Mozart-coder/BERT_winter-6_tokenized

sourceHugging Faceupdated 4y agoView on Hugging Face
0likes14downloads
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. -->

BERTwinter-6tokenized

This model is a fine-tuned version of armheb/DNA_bert_6 on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.0361

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: 2e-05
  • —trainbatchsize: 64
  • —evalbatchsize: 64
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 10
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation Loss
0.07061.02620.0369
0.03882.05240.0374
0.0373.07860.0363
0.03724.010480.0374
0.03765.013100.0351
0.03666.015720.0351
0.03777.018340.0373
0.03738.020960.0370
0.03589.023580.0359
0.036410.026200.0382

Framework versions

  • —Transformers 4.23.1
  • —Pytorch 1.12.1+cu113
  • —Datasets 2.6.1
  • —Tokenizers 0.13.1