metadata
license: apache-2.0
language:
- ko
- en
base_model:
- lcw99/t5-base-korean-text-summary
pipeline_tag: summarization
Model Description
lcw99๋์ด ๋ง๋์ lcw99/t5-base-korean-text-summary์ ๊ธฐ๋ฐ์ผ๋ก Finetuningํ์ฌ ๋ง๋ '๋ด์ค ๊ธฐ์ฌ ์์ฝ ๋ชจ๋ธ'์ ๋๋ค.
ํ์ต ๋ฐ์ดํฐ๋ AIHub์์ ์ ๊ณตํ๋ '๋ฌธ์์์ฝ ํ ์คํธ (https://aihub.or.kr/aihubdata/data/view.do?currMenu=115&topMenu=100&aihubDataSe=realm&dataSetSn=97)'์ ์ ๋ฌธ๊ธฐ์ฌ๋ค์ ์ฌ์ฉํ์์ต๋๋ค.
์ง์์ ์ผ๋ก ๋ ์ฑ๋ฅ์ ๊ฐ์ ํ์ฌ ์ข์ ๋ชจ๋ธ๋ก ๊ณ ๋ํ ํ๋๋ก ํ๊ฒ ์ต๋๋ค.
Training Arguments
training_args = Seq2SeqTrainingArguments(
evaluation_strategy="epoch",
save_strategy="epoch",
save_total_limit=2,
warmup_steps=1000,
learning_rate=5e-5,
per_device_train_batch_size=16,
per_device_eval_batch_size=16,
num_train_epochs=3,
weight_decay=0.01,
predict_with_generate=True,
fp16=True
)
Training Progress
Epoch | Training Loss | Validation Loss |
---|---|---|
1 | 0.604000 | 0.566043 |
2 | 0.577400 | 0.559071 |
3 | 0.553500 | 0.555571 |
์คํํ๊ฒฝ
Window 10
NVIDIA GeForce RTX 3070, 8192 MiB
Framework Versions
Python: 3.10.14
PyTorch: 1.12.1
Transformers: 4.46.2
Datasets: 3.2.0
Tokenizers: 0.20.3