metadata
license: apache-2.0
base_model: facebook/bart-base
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: emnlp_test_9clusters_msmarco
results: []
emnlp_test_9clusters_msmarco
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: 1.2540
- Bleu: 47.8433
- Gen Len: 10.199
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: 256
- eval_batch_size: 256
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| No log | 1.0 | 258 | 2.6072 | 17.3158 | 10.2598 |
| 3.0968 | 2.0 | 516 | 1.9501 | 29.2696 | 10.5689 |
| 3.0968 | 3.0 | 774 | 1.5769 | 38.3469 | 10.4586 |
| 1.911 | 4.0 | 1032 | 1.3777 | 44.1139 | 10.2272 |
| 1.911 | 5.0 | 1290 | 1.2820 | 46.7911 | 10.2506 |
| 1.5443 | 6.0 | 1548 | 1.2540 | 47.8433 | 10.199 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1