|
--- |
|
datasets: |
|
- mlsum |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: mukayese/transformer-turkish-summarization |
|
results: |
|
- task: |
|
name: Summarization |
|
type: summarization |
|
dataset: |
|
name: mlsum tu |
|
type: mlsum |
|
args: tu |
|
metrics: |
|
- name: Rouge1 |
|
type: rouge |
|
value: 43.2049 |
|
license: mit |
|
language: |
|
- tr |
|
pipeline_tag: summarization |
|
--- |
|
|
|
# [Mukayese: Turkish NLP Strikes Back](https://arxiv.org/abs/2203.01215) |
|
|
|
## Summarization: mukayese/transformer-turkish-summarization |
|
|
|
_This model is uncased_, it was initialized from scratch and trained only the mlsum/tu dataset with no pre-training. |
|
|
|
It achieves the following results on the evaluation set: |
|
|
|
- Rouge1: 43.2049 |
|
- Rouge2: 30.7082 |
|
- Rougel: 38.1981 |
|
- Rougelsum: 39.9453 |
|
|
|
Check [this](https://arxiv.org/abs/2203.01215) paper for more details on the model and the dataset. |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 0.0001 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 8 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 64 |
|
- total_eval_batch_size: 64 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 15.0 |
|
- mixed_precision_training: Native AMP |
|
- label_smoothing_factor: 0.1 |
|
|
|
### Framework versions |
|
|
|
- Transformers 4.11.3 |
|
- Pytorch 1.8.2+cu111 |
|
- Datasets 1.14.0 |
|
- Tokenizers 0.10.3 |
|
|
|
### Citation |
|
|
|
``` |
|
@misc{safaya-etal-2022-mukayese, |
|
title={Mukayese: Turkish NLP Strikes Back}, |
|
author={Ali Safaya and Emirhan Kurtuluş and Arda Göktoğan and Deniz Yuret}, |
|
year={2022}, |
|
eprint={2203.01215}, |
|
archivePrefix={arXiv}, |
|
primaryClass={cs.CL} |
|
} |
|
``` |
