metadata
license: mit
base_model: facebook/bart-large-cnn
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: bart-large-cnn-YT-transcript-sum
results: []
bart-large-cnn-YT-transcript-sum
This model is a fine-tuned version of facebook/bart-large-cnn on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.4333
- Rouge1: 51.3497
- Rouge2: 24.9492
- Rougel: 37.3016
- Rougelsum: 47.8317
- Gen Len: 97.6074
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: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 1.6217 | 1.0 | 540 | 1.4424 | 50.4219 | 24.0335 | 35.9464 | 46.8506 | 90.2630 |
| 1.0592 | 2.0 | 1080 | 1.4333 | 51.3497 | 24.9492 | 37.3016 | 47.8317 | 97.6074 |
| 0.723 | 3.0 | 1620 | 1.6339 | 50.9967 | 24.488 | 37.1713 | 47.2127 | 98.6370 |
| 0.4971 | 4.0 | 2160 | 1.7243 | 51.2797 | 24.0696 | 36.6573 | 47.4137 | 99.8370 |
| 0.3559 | 5.0 | 2700 | 1.9144 | 51.4203 | 24.3533 | 36.8606 | 47.6179 | 96.4481 |
| 0.2528 | 6.0 | 3240 | 2.1977 | 51.3129 | 24.724 | 36.992 | 47.8757 | 97.7630 |
| 0.1804 | 7.0 | 3780 | 2.3279 | 51.3644 | 23.9238 | 36.5924 | 47.2068 | 94.4481 |
| 0.1331 | 8.0 | 4320 | 2.4088 | 51.742 | 24.7768 | 37.6327 | 48.0398 | 90.7778 |
| 0.0949 | 9.0 | 4860 | 2.5075 | 52.0502 | 24.8043 | 37.5371 | 48.0676 | 86.2037 |
| 0.074 | 10.0 | 5400 | 2.5641 | 52.0733 | 25.2822 | 37.6324 | 48.3677 | 93.3370 |
| 0.0506 | 11.0 | 5940 | 2.7945 | 52.2919 | 25.5404 | 37.956 | 48.4453 | 93.2222 |
| 0.0343 | 12.0 | 6480 | 2.8614 | 52.0782 | 25.039 | 37.5952 | 48.469 | 89.5222 |
| 0.0126 | 13.0 | 7020 | 3.0071 | 52.3343 | 25.4229 | 37.8453 | 48.4345 | 92.5889 |
| 0.0065 | 14.0 | 7560 | 3.0398 | 52.0812 | 24.9641 | 37.409 | 48.2338 | 94.3037 |
| 0.0032 | 15.0 | 8100 | 3.0438 | 52.184 | 25.1639 | 37.4424 | 48.3197 | 93.2333 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3