<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
mbart-large-50-pluska-token-sum
This model is a fine-tuned version of facebook/mbart-large-50 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 4.2944
- Rouge1: 16.8143
- Rouge2: 5.2102
- Rougel: 14.3454
- Rougelsum: 14.7355
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: 5.6e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 144 | 3.7149 | 17.1082 | 5.3773 | 14.6384 | 14.9023 |
| No log | 2.0 | 288 | 3.4301 | 17.6442 | 5.7235 | 14.9728 | 15.3384 |
| No log | 3.0 | 432 | 3.4951 | 17.491 | 5.5463 | 14.8284 | 15.2035 |
| 2.9135 | 4.0 | 576 | 3.6706 | 17.1471 | 5.3419 | 14.6302 | 14.9667 |
| 2.9135 | 5.0 | 720 | 3.9122 | 16.6979 | 5.1895 | 14.2909 | 14.6367 |
| 2.9135 | 6.0 | 864 | 4.0815 | 16.5961 | 5.1915 | 14.2459 | 14.5916 |
| 2.9135 | 7.0 | 1008 | 4.2448 | 16.7793 | 5.2088 | 14.3515 | 14.7136 |
| 1.0097 | 8.0 | 1152 | 4.2944 | 16.8143 | 5.2102 | 14.3454 | 14.7355 |
Framework versions
- Transformers 4.27.0
- Pytorch 2.0.0
- Datasets 2.10.1
- Tokenizers 0.13.2
