roberta-base-infringement-detect
Model Details
Model Description
klue/roberta-base ๋ชจ๋ธ์ ์ด์ฉํ์ฌ, ๋ ์ปจํ ์ธ ๊ฐ์ ์ ์ฌ์ฌ๋ถ๋ฅผ ํ์ธํ๋ ๋ชจ๋ธ์ ๋๋ค.
Train
์์ฒด๊ตฌ์ถ๋ 1,310๊ฐ์ ์ฐธ์ธ ์ ์ฌ ์ปจํ ์ธ ์์ ์ด์ฉํ์ฌ, ์ ํ ํ ์ฐธ/๊ฑฐ์ง ๋น์จ 1:2์ธ ๋ฐ์ดํฐ์ ์ ์์ฑํ์ฌ ํ์ต์์ผฐ์ต๋๋ค.
์ด์ธ์ ํ์ต์ ํ๋ผ๋ฏธํฐ๋ ๋ค์๊ณผ ๊ฐ์ต๋๋ค.
Parameter | Value |
---|---|
train_batch_size |
16 |
num_train_epochs |
5 |
weight_decay |
0.01 |
learning_rate |
2e-5 |
How to use
from transformers import AutoTokenizer, AutoModelForSequenceClassification
model_name = "kms7530/roberta-base-infringement-detect"
model = AutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
๋ชจ๋ธ์ ์ถ๋ก ์ ๋ค์๊ณผ ๊ฐ์ด ์ ๋ ฅํด์ผ ํฉ๋๋ค.
[CLS]\
[unused0]<ORIGINAL_CONTENT_TITLE>\
[unused1]<ORIGINAL_CONTENT>[SEP] \
[unused0]<TEST_CONTENT_TITLE>\
[unused1]<TEST_CONTENT>[SEP]
- Downloads last month
- 6
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.