metadata
license: mit
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: mbart-large-50-finetuned-sah-to-feat
results: []
mbart-large-50-finetuned-sah-to-feat
This model is a fine-tuned version of facebook/mbart-large-50 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3669
- Bleu: 41.8685
- Gen Len: 18.2677
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: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
---|---|---|---|---|---|
No log | 1.0 | 24 | 2.6306 | 23.9912 | 34.3858 |
No log | 2.0 | 48 | 1.9570 | 39.8674 | 22.4882 |
No log | 3.0 | 72 | 1.5854 | 39.2001 | 21.1102 |
No log | 4.0 | 96 | 0.9981 | 26.199 | 31.5984 |
No log | 5.0 | 120 | 0.4687 | 42.9123 | 20.6614 |
No log | 6.0 | 144 | 0.4401 | 39.42 | 19.7244 |
No log | 7.0 | 168 | 0.3543 | 48.7148 | 21.1732 |
No log | 8.0 | 192 | 0.3650 | 46.2897 | 20.9843 |
No log | 9.0 | 216 | 0.3539 | 41.6928 | 18.4567 |
No log | 10.0 | 240 | 0.3669 | 41.8685 | 18.2677 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1
- Datasets 2.13.1
- Tokenizers 0.13.3