agi-css's picture
update model card README.md
bbaca24
|
raw
history blame
1.83 kB
metadata
license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: distilroberta-base-etc
    results: []

distilroberta-base-etc

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

  • Loss: 0.5308
  • Accuracy: 0.7764
  • F1: 0.7744

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: 1.786634543080922e-05
  • train_batch_size: 400
  • eval_batch_size: 400
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 7

Training results

Training Loss Epoch Step Validation Loss Accuracy F1
No log 1.0 46 0.6594 0.6699 0.6581
No log 2.0 92 0.5253 0.7575 0.7574
No log 3.0 138 0.5651 0.7394 0.7338
No log 4.0 184 0.4815 0.7862 0.7860
No log 5.0 230 0.5280 0.7734 0.7721
No log 6.0 276 0.5066 0.7786 0.7776
No log 7.0 322 0.5308 0.7764 0.7744

Framework versions

  • Transformers 4.18.0
  • Pytorch 1.11.0+cu113
  • Datasets 2.1.0
  • Tokenizers 0.12.1