<!-- 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. -->
ctrlsum-cnndm-last-500-all
This model is a fine-tuned version of hyunwoongko/ctrlsum-cnndm on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.3405
- Rouge1: 47.0674
- Rouge2: 32.3086
- Rougel: 43.1016
- Rougelsum: 44.0044
- Gen Len: 16.5870
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: 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 | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 191 | 1.3925 | 42.611 | 28.7644 | 38.7719 | 39.5325 | 16.9027 |
| No log | 2.0 | 382 | 1.3405 | 47.0674 | 32.3086 | 43.1016 | 44.0044 | 16.5870 |
| 1.2963 | 3.0 | 573 | 1.3645 | 46.8248 | 31.943 | 42.7128 | 43.3736 | 16.8702 |
| 1.2963 | 4.0 | 764 | 1.3977 | 47.6883 | 31.4445 | 43.2304 | 44.1072 | 16.8112 |
| 1.2963 | 5.0 | 955 | 1.4410 | 47.7499 | 31.7408 | 43.3394 | 44.1869 | 17.4071 |
| 0.7 | 6.0 | 1146 | 1.5007 | 47.8986 | 31.9365 | 43.3637 | 44.2524 | 16.8702 |
| 0.7 | 7.0 | 1337 | 1.6100 | 47.7646 | 31.4891 | 43.1422 | 43.9226 | 17.7168 |
| 0.4263 | 8.0 | 1528 | 1.6834 | 47.9059 | 31.5554 | 43.436 | 44.2052 | 17.1327 |
| 0.4263 | 9.0 | 1719 | 1.7464 | 47.6552 | 31.3669 | 43.0738 | 43.8913 | 16.7994 |
| 0.4263 | 10.0 | 1910 | 1.8040 | 47.4039 | 30.9281 | 42.8384 | 43.6918 | 17.0118 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3
