anzeo's picture
loha_fine_tuned_rte_XLMroberta
9a184d6 verified
metadata
license: mit
library_name: peft
tags:
  - generated_from_trainer
base_model: xlm-roberta-base
metrics:
  - accuracy
  - f1
model-index:
  - name: loha_fine_tuned_rte_XLMroberta
    results: []

loha_fine_tuned_rte_XLMroberta

This model is a fine-tuned version of xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 3.0980
  • Accuracy: 0.6207
  • F1: 0.6090

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: 0.003
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • training_steps: 400

Training results

Training Loss Epoch Step Validation Loss Accuracy F1
0.8165 1.7241 50 0.7174 0.4828 0.3781
0.7386 3.4483 100 0.6616 0.6897 0.6523
0.7293 5.1724 150 0.7683 0.5172 0.4660
0.6773 6.8966 200 1.1129 0.4483 0.4324
0.4623 8.6207 250 1.7863 0.5862 0.5892
0.2532 10.3448 300 2.8440 0.5862 0.5483
0.0813 12.0690 350 3.0842 0.5517 0.5484
0.0478 13.7931 400 3.0980 0.6207 0.6090

Framework versions

  • PEFT 0.11.1
  • Transformers 4.40.2
  • Pytorch 2.1.1+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1