<!-- 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. -->
final-squad-bn-qgen-banglat5-all-metric-v3
This model is a fine-tuned version of csebuetnlp/banglat5 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4231
- Rouge1 Precision: 39.5263
- Rouge1 Recall: 37.168
- Rouge1 Fmeasure: 37.1806
- Rouge2 Precision: 17.7035
- Rouge2 Recall: 16.508
- Rouge2 Fmeasure: 16.5336
- Rougel Precision: 37.135
- Rougel Recall: 34.9177
- Rougel Fmeasure: 34.9266
- Rougelsum Precision: 37.1205
- Rougelsum Recall: 34.8982
- Rougelsum Fmeasure: 34.9129
- Bleu-1: 36.4356
- Bleu-2: 22.3217
- Bleu-3: 14.7682
- Bleu-4: 10.0865
- Meteor: 0.2051
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
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 | Bleu-1 | Bleu-2 | Bleu-3 | Bleu-4 | Meteor |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.5901 | 1.0 | 6769 | 0.4756 | 32.2563 | 31.3211 | 30.7652 | 12.2914 | 11.9567 | 11.6739 | 29.1321 | 28.3925 | 27.84 | 29.1291 | 28.3832 | 27.8378 | 31.9366 | 17.9528 | 10.9479 | 6.8658 | 0.1715 |
| 0.5094 | 2.0 | 13538 | 0.4343 | 37.5727 | 35.6711 | 35.4661 | 16.3104 | 15.4196 | 15.3046 | 35.2059 | 33.4452 | 33.2559 | 35.1882 | 33.4395 | 33.24 | 35.1532 | 21.1183 | 13.7297 | 9.2128 | 0.1955 |
| 0.4866 | 3.0 | 20307 | 0.4267 | 38.6402 | 36.2947 | 36.2796 | 16.8569 | 15.7129 | 15.7114 | 36.2902 | 34.0855 | 34.0734 | 36.2733 | 34.0723 | 34.0661 | 35.6286 | 21.554 | 14.098 | 9.5506 | 0.1996 |
| 0.4732 | 4.0 | 27076 | 0.4235 | 39.3469 | 36.7357 | 36.8598 | 17.4835 | 16.2062 | 16.2677 | 36.9883 | 34.5422 | 34.6543 | 36.9783 | 34.5352 | 34.6594 | 35.9917 | 21.9745 | 14.4922 | 9.884 | 0.203 |
| 0.4646 | 5.0 | 33845 | 0.4224 | 39.4223 | 37.0956 | 37.0893 | 17.6277 | 16.4682 | 16.4692 | 37.0994 | 34.896 | 34.8991 | 37.0885 | 34.8691 | 34.8811 | 36.3637 | 22.2704 | 14.7068 | 10.021 | 0.2049 |
| 0.4517 | 6.0 | 40614 | 0.4231 | 39.5263 | 37.168 | 37.1806 | 17.7035 | 16.508 | 16.5336 | 37.135 | 34.9177 | 34.9266 | 37.1205 | 34.8982 | 34.9129 | 36.4356 | 22.3217 | 14.7682 | 10.0865 | 0.2051 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1
