Shijia's picture
End of training
22fa56c verified
metadata
license: mit
base_model: FacebookAI/xlm-roberta-base
tags:
  - generated_from_trainer
model-index:
  - name: xlm-roberta-base_afr_corr_0.0001
    results: []

xlm-roberta-base_afr_corr_0.0001

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

  • Loss: 0.0461
  • Spearman Corr: -0.0407

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: 32
  • eval_batch_size: 128
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 30
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Spearman Corr
No log 0.85 200 0.0466 0.1403
No log 1.69 400 0.0463 nan
0.0564 2.54 600 0.0476 -0.0126
0.0564 3.38 800 0.0463 nan
0.0519 4.23 1000 0.0481 -0.0333
0.0519 5.07 1200 0.0463 -0.1378
0.0519 5.92 1400 0.0480 -0.0688
0.051 6.77 1600 0.0463 nan
0.051 7.61 1800 0.0461 -0.0407

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.2.0+cu121
  • Datasets 2.17.0
  • Tokenizers 0.15.2