Edit model card

bigbird-roberta-large

This model is a fine-tuned version of LIAMF-USP/roberta-large-finetuned-race on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.6094
  • Accuracy: 0.1976
  • F1: 0.1757
  • Precision: 0.1893
  • Recall: 0.1911

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: 2e-05
  • train_batch_size: 8
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall
1.6272 0.3233 1200 1.6094 0.2082 0.1431 0.2007 0.1996
1.6218 0.6466 2400 1.6094 0.2117 0.1340 0.1876 0.1998
1.6235 0.9698 3600 1.6094 0.2104 0.1752 0.2005 0.2015
1.617 1.2931 4800 1.6094 0.2088 0.1956 0.2037 0.2028
1.61 1.6164 6000 1.6094 0.2091 0.1606 0.2127 0.2024
1.6126 1.9397 7200 1.6094 0.2108 0.1796 0.1965 0.2011
1.6174 2.2629 8400 1.6094 0.2095 0.1833 0.2036 0.2024
1.6125 2.5862 9600 1.6094 0.2097 0.1847 0.1963 0.2016
1.6192 2.9095 10800 1.6094 0.1976 0.1757 0.1893 0.1911

Framework versions

  • Transformers 4.40.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
0
Safetensors
Model size
355M params
Tensor type
F32
·
Inference API (serverless) does not yet support transformers models for this pipeline type.

Finetuned from