| | --- |
| | license: apache-2.0 |
| | tags: |
| | - generated_from_trainer |
| | metrics: |
| | - rouge |
| | model-index: |
| | - name: mt5-summarize-sum |
| | results: [] |
| | --- |
| | |
| | <!-- 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. --> |
| |
|
| | # mt5-summarize-sum |
| |
|
| | This model is a fine-tuned version of [google/mt5-base](https://huggingface.co/google/mt5-base) on the None dataset. |
| | It achieves the following results on the evaluation set: |
| | - Loss: 0.3984 |
| | - Rouge1: 0.5736 |
| | - Rouge2: 0.3783 |
| | - Rougel: 0.4855 |
| | - Rougelsum: 0.4844 |
| |
|
| | ## 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: 0.0005 |
| | - train_batch_size: 2 |
| | - eval_batch_size: 2 |
| | - seed: 42 |
| | - gradient_accumulation_steps: 16 |
| | - total_train_batch_size: 32 |
| | - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
| | - lr_scheduler_type: linear |
| | - lr_scheduler_warmup_steps: 90 |
| | - num_epochs: 3 |
| |
|
| | ### Training results |
| |
|
| | | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | |
| | |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:| |
| | | 13.8551 | 0.16 | 100 | 5.4672 | 0.2389 | 0.0546 | 0.2119 | 0.2110 | |
| | | 1.0762 | 0.33 | 200 | 0.5982 | 0.3774 | 0.2199 | 0.3493 | 0.3470 | |
| | | 0.8077 | 0.49 | 300 | 0.4999 | 0.4929 | 0.3195 | 0.4349 | 0.4312 | |
| | | 0.7772 | 0.65 | 400 | 0.4652 | 0.4715 | 0.3296 | 0.4431 | 0.4409 | |
| | | 0.7771 | 0.82 | 500 | 0.4402 | 0.4881 | 0.3356 | 0.4433 | 0.4412 | |
| | | 0.713 | 0.98 | 600 | 0.4500 | 0.4990 | 0.3291 | 0.4550 | 0.4525 | |
| | | 0.65 | 1.15 | 700 | 0.4335 | 0.5522 | 0.3633 | 0.4930 | 0.4909 | |
| | | 0.7035 | 1.31 | 800 | 0.4278 | 0.5227 | 0.3470 | 0.4781 | 0.4772 | |
| | | 0.6818 | 1.47 | 900 | 0.4202 | 0.5325 | 0.3585 | 0.4759 | 0.4744 | |
| | | 0.6643 | 1.64 | 1000 | 0.4113 | 0.5326 | 0.3486 | 0.4678 | 0.4641 | |
| | | 0.6007 | 1.8 | 1100 | 0.4122 | 0.5152 | 0.3260 | 0.4572 | 0.4547 | |
| | | 0.5866 | 1.96 | 1200 | 0.4158 | 0.5538 | 0.3680 | 0.4910 | 0.4903 | |
| | | 0.5563 | 2.13 | 1300 | 0.4051 | 0.5433 | 0.3371 | 0.4685 | 0.4672 | |
| | | 0.5727 | 2.29 | 1400 | 0.4089 | 0.5447 | 0.3619 | 0.4711 | 0.4695 | |
| | | 0.5859 | 2.45 | 1500 | 0.4033 | 0.5464 | 0.3411 | 0.4688 | 0.4662 | |
| | | 0.5783 | 2.62 | 1600 | 0.3997 | 0.5667 | 0.3595 | 0.4825 | 0.4787 | |
| | | 0.5673 | 2.78 | 1700 | 0.3992 | 0.5759 | 0.3882 | 0.4911 | 0.4891 | |
| | | 0.57 | 2.95 | 1800 | 0.3984 | 0.5736 | 0.3783 | 0.4855 | 0.4844 | |
| |
|
| |
|
| | ### Framework versions |
| |
|
| | - Transformers 4.27.4 |
| | - Pytorch 1.13.0 |
| | - Datasets 2.1.0 |
| | - Tokenizers 0.13.2 |
| |
|
