CoolFace
Modelpublic

irfanamal/distilroberta_science_multiple_choice

sourceHugging Faceapache-2.0updated 3y agoView on Hugging Face
0likes17downloads
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. -->

distilrobertasciencemultiple_choice

This model is a fine-tuned version of irfanamal/distilroberta_multiple_choice on the None dataset. It achieves the following results on the evaluation set:

  • —Loss: 1.1923
  • —Accuracy: 0.45

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-06
  • —trainbatchsize: 16
  • —evalbatchsize: 16
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 100

Training results

Training LossEpochStepValidation LossAccuracy
1.25891.0121.19570.55
1.28562.0241.19130.45
1.24563.0361.19080.5
1.17534.0481.18880.45
1.09415.0601.18500.45
1.07956.0721.17640.45
1.02427.0841.16550.45
0.95768.0961.15580.45
0.92719.01081.16040.45
0.904710.01201.17480.45
0.902511.01321.17830.45
0.78212.01441.17870.45
0.800313.01561.19230.45

Framework versions

  • —Transformers 4.31.0
  • —Pytorch 2.0.1+cu117
  • —Datasets 2.14.4
  • —Tokenizers 0.13.3