metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
- generated_from_trainer
model-index:
- name: pubmed-mixed-noise-v3-0.4
results: []
pubmed-mixed-noise-v3-0.4
This model is a fine-tuned version of facebook/bart-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7276
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.4148 | 0.11 | 500 | 1.2099 |
| 1.2472 | 0.21 | 1000 | 1.0718 |
| 1.0935 | 0.32 | 1500 | 0.9924 |
| 1.0193 | 0.43 | 2000 | 0.9482 |
| 0.8859 | 0.54 | 2500 | 0.9335 |
| 0.9838 | 0.64 | 3000 | 0.8792 |
| 0.9195 | 0.75 | 3500 | 0.8665 |
| 0.9793 | 0.86 | 4000 | 0.8397 |
| 0.8513 | 0.96 | 4500 | 0.8215 |
| 0.7384 | 1.07 | 5000 | 0.8261 |
| 0.8117 | 1.18 | 5500 | 0.8028 |
| 0.8606 | 1.28 | 6000 | 0.7930 |
| 0.8139 | 1.39 | 6500 | 0.7873 |
| 0.84 | 1.5 | 7000 | 0.7761 |
| 0.733 | 1.61 | 7500 | 0.7728 |
| 0.7102 | 1.71 | 8000 | 0.7644 |
| 0.6857 | 1.82 | 8500 | 0.7579 |
| 0.7244 | 1.93 | 9000 | 0.7500 |
| 0.5931 | 2.03 | 9500 | 0.7515 |
| 0.6165 | 2.14 | 10000 | 0.7484 |
| 0.6681 | 2.25 | 10500 | 0.7414 |
| 0.6716 | 2.35 | 11000 | 0.7390 |
| 0.6468 | 2.46 | 11500 | 0.7387 |
| 0.5505 | 2.57 | 12000 | 0.7361 |
| 0.635 | 2.68 | 12500 | 0.7333 |
| 0.6632 | 2.78 | 13000 | 0.7297 |
| 0.6271 | 2.89 | 13500 | 0.7278 |
| 0.5843 | 3.0 | 14000 | 0.7276 |
Framework versions
- Transformers 4.36.1
- Pytorch 2.0.1
- Datasets 2.16.1
- Tokenizers 0.15.0