metadata
license: apache-2.0
base_model: facebook/bart-large
tags:
- generated_from_trainer
metrics:
- rouge
- wer
model-index:
- name: BART_1st_STAGE_SUMMARIZER
results: []
BART_1st_STAGE_SUMMARIZER
This model is a fine-tuned version of facebook/bart-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1826
- Rouge1: 0.7384
- Rouge2: 0.5134
- Rougel: 0.6809
- Rougelsum: 0.6852
- Wer: 0.3923
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: 6
- eval_batch_size: 6
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Wer |
|---|---|---|---|---|---|---|---|---|
| No log | 0.21 | 250 | 1.3973 | 0.7211 | 0.486 | 0.6611 | 0.6661 | 0.417 |
| 1.969 | 0.42 | 500 | 1.2499 | 0.7303 | 0.4988 | 0.67 | 0.6745 | 0.4056 |
| 1.969 | 0.63 | 750 | 1.2039 | 0.734 | 0.5068 | 0.6761 | 0.6798 | 0.3977 |
| 1.3659 | 0.84 | 1000 | 1.1826 | 0.7384 | 0.5134 | 0.6809 | 0.6852 | 0.3923 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2