<!-- 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. -->
squad-bn-qgen-banglat5
This model is a fine-tuned version of csebuetnlp/banglat5 on the squad_bn dataset. It achieves the following results on the evaluation set:
- Loss: 0.4808
- Rouge1 Precision: 37.7366
- Rouge1 Recall: 34.2712
- Rouge1 Fmeasure: 34.8738
- Rouge2 Precision: 16.2055
- Rouge2 Recall: 14.568
- Rouge2 Fmeasure: 14.852
- Rougel Precision: 35.4241
- Rougel Recall: 32.2011
- Rougel Fmeasure: 32.7617
- Rougelsum Precision: 35.4167
- Rougelsum Recall: 32.1978
- Rougelsum Fmeasure: 32.7572
- Sacrebleu: 8.0898
- Meteor: 0.1782
- Gen Len: 9.8299
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: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 Precision | Rouge1 Recall | Rouge1 Fmeasure | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure | Rougel Precision | Rougel Recall | Rougel Fmeasure | Rougelsum Precision | Rougelsum Recall | Rougelsum Fmeasure | Sacrebleu | Meteor | Gen Len |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.5208 | 1.0 | 16396 | 0.4683 | 38.566 | 35.5094 | 35.9216 | 17.0701 | 15.3916 | 15.6829 | 36.4433 | 33.5298 | 33.958 | 36.4637 | 33.5496 | 33.9913 | 8.6055 | 0.1799 | 9.8340 |
| 0.479 | 2.0 | 32792 | 0.4815 | 40.7475 | 35.8163 | 37.0498 | 17.9002 | 15.2742 | 15.9601 | 38.6977 | 33.8607 | 35.1258 | 38.7261 | 33.8717 | 35.1537 | 9.0561 | 0.1835 | 9.4338 |
| 0.4577 | 3.0 | 49188 | 0.4879 | 40.6712 | 36.2763 | 37.2775 | 18.5942 | 16.0689 | 16.7206 | 38.8546 | 34.5013 | 35.5491 | 38.8633 | 34.5255 | 35.5682 | 9.7947 | 0.1879 | 9.6324 |
| 0.4389 | 4.0 | 65584 | 0.4881 | 41.4251 | 36.2873 | 37.6272 | 18.561 | 15.7067 | 16.5358 | 39.434 | 34.3496 | 35.7457 | 39.533 | 34.4702 | 35.8347 | 9.7612 | 0.1881 | 9.3944 |
| 0.4321 | 5.0 | 81980 | 0.4937 | 41.1197 | 36.0568 | 37.4121 | 18.7179 | 15.8348 | 16.6644 | 39.3386 | 34.3177 | 35.7088 | 39.3171 | 34.3015 | 35.6748 | 9.8263 | 0.1887 | 9.4040 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1
