baseline / README.md
Yova's picture
End of training
3d34026
|
raw
history blame
2.58 kB
metadata
tags:
  - generated_from_trainer
model-index:
  - name: baseline
    results: []

baseline

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

  • Loss: 1.2583
  • Exact Match: 0.0

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: 0.001
  • train_batch_size: 100
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 400
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: inverse_sqrt
  • lr_scheduler_warmup_steps: 4000
  • num_epochs: 20

Training results

Training Loss Epoch Step Validation Loss Exact Match
6.6009 1.0 25 5.6681 0.0
5.9582 2.0 50 4.4844 0.0
5.0101 3.0 75 3.7434 0.0
4.3172 4.0 100 3.4055 0.0
3.9672 5.0 125 3.2224 0.0
3.7659 6.0 150 3.0110 0.0
3.5416 7.0 175 2.7741 0.0
3.3002 8.0 200 2.4967 0.0
3.0412 9.0 225 2.2354 0.0
2.8117 10.0 250 2.1391 0.0
2.6168 11.0 275 2.0822 0.0
2.4624 12.0 300 2.0147 0.0
2.3253 13.0 325 1.9378 0.0
2.2152 14.0 350 1.8335 0.0
2.11 15.0 375 1.7586 0.0
2.0029 16.0 400 1.6847 0.0
1.9103 17.0 425 1.5874 0.0
1.8166 18.0 450 1.5552 0.0
1.7264 19.0 475 1.3748 0.0
1.6562 20.0 500 1.2583 0.0

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu118
  • Datasets 2.15.0
  • Tokenizers 0.15.0