Edit model card

finetuned-base_mini

This model is a fine-tuned version of google/bert_uncased_L-4_H-256_A-4 on the imdb dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3938
  • Accuracy: 0.9076
  • F1: 0.9516

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: 3e-05
  • train_batch_size: 128
  • eval_batch_size: 128
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: constant
  • num_epochs: 200

Training results

Training Loss Epoch Step Validation Loss Accuracy F1
0.354 2.55 500 0.2300 0.9116 0.9538
0.2086 5.1 1000 0.3182 0.8815 0.9370
0.1401 7.65 1500 0.2160 0.9241 0.9605
0.0902 10.2 2000 0.4684 0.8722 0.9317
0.0654 12.76 2500 0.4885 0.8747 0.9332
0.043 15.31 3000 0.3938 0.9076 0.9516

Framework versions

  • Transformers 4.25.0
  • Pytorch 1.12.1+cu113
  • Datasets 2.7.1
  • Tokenizers 0.13.2
Downloads last month
2

Finetuned from

Dataset used to train muhtasham/finetuned-base_mini

Evaluation results