joelniklaus's picture
update model card README.md
5035f3d
metadata
tags:
  - generated_from_trainer
model-index:
  - name: legal-irish-roberta-base
    results: []

legal-irish-roberta-base

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7328

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: 0.0001
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • distributed_type: tpu
  • num_devices: 8
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 512
  • total_eval_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.05
  • training_steps: 200000

Training results

Training Loss Epoch Step Validation Loss
0.5892 228.0 50000 0.7659
0.4497 456.0 100000 0.7421
0.3906 684.0 150000 0.7443
0.3906 913.0 200000 0.7328

Framework versions

  • Transformers 4.20.1
  • Pytorch 1.12.0+cu102
  • Datasets 2.9.0
  • Tokenizers 0.12.0