metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: bart-finetuned-pubmed
results: []
bart-finetuned-pubmed
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: 1.7626
- Rouge2 Precision: 0.3149
- Rouge2 Recall: 0.2118
- Rouge2 Fmeasure: 0.2349
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: 16
- 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
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |
---|---|---|---|---|---|---|
No log | 1.0 | 100 | 1.7898 | 0.3222 | 0.2373 | 0.2535 |
No log | 2.0 | 200 | 1.7738 | 0.3125 | 0.2167 | 0.2402 |
No log | 3.0 | 300 | 1.7626 | 0.3149 | 0.2118 | 0.2349 |
Framework versions
- Transformers 4.12.2
- Pytorch 1.9.0+cu111
- Datasets 1.14.0
- Tokenizers 0.10.3