Wangchanberta Base Wiki 20210520 News Spm Finetune Qa

🚀 wangchanberta-base-wiki-20210520-news-spm-finetune-qa

This project focuses on fine - tuning the airesearchth/wangchanberta-base-wiki-20210520-news-spm model. It uses the training sets from iapp_wiki_qa_squad, thaiqa_squad, and nsc_qa. For the latter two, examples with a cosine similarity of over 0.8 to the validation and test examples are removed, and their contexts are trimmed to around 300 newmm words. Benchmarks are shared on wandb using the validation and test sets of iapp_wiki_qa_squad. The model is trained with thai2transformers.

🚀 Quick Start

📦 Installation

There is no specific installation content provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

To run the fine - tuning process, use the following commands:

export MODEL_NAME=airesearchth/wangchanberta-base-wiki-20210520-news-spm
CUDA_LAUNCH_BLOCKING=1 python train_question_answering_lm_finetuning.py \
  --model_name $MODEL_NAME \
  --dataset_name chimera_qa \
  --output_dir $MODEL_NAME-finetune-chimera_qa-model \
  --log_dir $MODEL_NAME-finetune-chimera_qa-log \
  --model_max_length 400 \
  --pad_on_right \
  --fp16

🔧 Technical Details

There is no specific technical details content provided in the original document, so this section is skipped.

📄 License

There is no license information provided in the original document, so this section is skipped.