Instructions to use boostcamp-5th-nlp07/klue-roberta-large-reader-noNewline with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use boostcamp-5th-nlp07/klue-roberta-large-reader-noNewline with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("question-answering", model="boostcamp-5th-nlp07/klue-roberta-large-reader-noNewline")# Load model directly from transformers import AutoTokenizer, AutoModelForQuestionAnswering tokenizer = AutoTokenizer.from_pretrained("boostcamp-5th-nlp07/klue-roberta-large-reader-noNewline") model = AutoModelForQuestionAnswering.from_pretrained("boostcamp-5th-nlp07/klue-roberta-large-reader-noNewline", device_map="auto") - Notebooks
- Google Colab
- Kaggle
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
설명
'\n'을 제거한 데이터셋으로 학습시킨 모델.
사용한 arguments
python train.py --output_dir ${MODEL_PATH} \
--wandb_name ${WANDB_NAME} \
--do_train --do_eval \
--model_name_or_path "klue/roberta-large" \
--dataset_name ../data/noNewline_train_dataset \
--save_strategy steps \
--save_steps 500 \
--save_total_limit 1 \
--logging_steps 100 \
--evaluation_strategy steps \
--eval_steps 500 \
--per_device_train_batch_size 8 \
--per_device_eval_batch_size 8 \
--learning_rate 5e-5 \
--weight_decay 0.0 \
--num_train_epochs 3.0 \
--lr_scheduler_type linear \
--warmup_ratio 0.0 \
--report_to wandb
- Downloads last month
- 4