license: cc-by-nc-sa-4.0
tags:
- generated_from_trainer
- simplification
task_categories:
- text2text-generation
task_ids:
- text-simplification
language:
- nl
datasets:
- BramVanroy/chatgpt-dutch-simplification
metrics:
- rouge
- sari
model-index:
- name: BramVanroy/ul2-large-dutch-simplification-mai-2023
results:
- task:
type: text-simplification
name: Text Simplification
dataset:
type: BramVanroy/chatgpt-dutch-simplification
name: ChatGPT Dutch Simplification
metrics:
- type: rouge
value: 41.3871
name: Eval Rouge-1
- type: rouge
value: 19.6751
name: Eval Rouge-2
- type: rouge
value: 36.0469
name: Eval RougeL
- type: rouge
value: 36.1178
name: Eval RougeLsum
- type: sari
value: 54.3588
name: Eval SARI
- type: rouge
value: 43.8191
name: Test Rouge-1
- type: rouge
value: 21.7783
name: Test Rouge-2
- type: rouge
value: 39.3657
name: Test RougeL
- type: rouge
value: 39.3751
name: Test RougeLsum
- type: sari
value: 52.3752
name: Test SARI
widget:
- example_title: Cooking
text: >-
Op bepaalde tijdstippen verlang ik naar de smaakvolle culinaire creaties
welke door de ambachtelijke expertise van mijn grootmoeder zijn
vervaardigd.
ul2-large-dutch-simplification-mai-2023
This model is intended to simplify Dutch sentences.
This model is a fine-tuned version of yhavinga/ul2-large-dutch on the BramVanroy/chatgpt-dutch-simplification dataset.
The model was created in light of the master thesis of Charlotte Van de Velde in the Master of Science in Artificial Intelligence (MAI) at KU Leuven in 2023. Charlotte is supervised by Vincent Vandeghinste and Bram Vanroy. Dataset creation by Charlotte, model training by Bram.
Quick links
- Repository: includes training code and model creation log
- Dataset:
BramVanroy/chatgpt-dutch-simplification - Parent model: this model was finetuned on
yhavinga/ul2-large-dutch - Demo: shows the "base" model in action (don't rely on the "Hosted inference API" widget on this page, it does not work very well)
Intended uses & limitations, and dataset
The model is intended for sentence-level simplification of Dutch. It might extend to document-level simplification but most of the dataset is limited to sentences so document-level performance is not guaranteed.
The dataset has been generated automatically (cf. dataset description) and has not been manually verified. On top of that, this model has been fine-tuned and we did not scrutinize the parent model or its training data. Output of the current model is therefore subject to unexpected results (as most if not all neural networks).
Because the dataset was generated with ChatGPT, this model cannot be used for commercial purposes.
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002927210895006501
- train_batch_size: 32
- optimizer: Adafactor
- num_epochs: 27
These hyperarameters were found through Bayesian hyperparameter search with wandb. This is described in the
repository.
Training results
eval results are on the evaluation set, predict results are on the test set. These were achieved with
beam search (num_beams=3).
{
"eval_gen_len": 21.404761904761905,
"eval_loss": 3.0882697105407715,
"eval_rouge1": 41.3871,
"eval_rouge2": 19.6751,
"eval_rougeL": 36.0469,
"eval_rougeLsum": 36.1178,
"eval_sari": 54.3588,
"predict_gen_len": 22.1484375,
"predict_loss": 2.7822625637054443,
"predict_rouge1": 43.8191,
"predict_rouge2": 21.7783,
"predict_rougeL": 39.3657,
"predict_rougeLsum": 39.3751,
"predict_sari": 52.3752
}
Note: the model seems to underperform compared to the base variant of the model, achieving only similar results with a much larger size. The reason for this may be found in the hyperparameters, where this large model may have benefitted from a smaller learning rate in the optimisation space. In the hyperparameter search, the learning rate spectrum was set to 1e-03 to 1e-04 but this might be too large for this model and size.
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3