Downloads · 30 days
7
0% of all-time downloads
kor-3i4k-bert-base-cased
seongju/kor-3i4k-bert-base-cased
kor-3i4k-bert-base-cased is a text classification model from seongju. Use it when you need a label for a piece of text. It is set up for transformers.
language : Korean fine tuning data : kor3i4k License : CC-BY-SA 4.0 Base model : bert-base-multilingual-cased input : sentence output : intent
All-time downloads
1.5K
Public
Repo size
1.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin712 MB · 100%
At a glance
- Task
- Text Classification
- Library
- transformers
- Model type
- bert
- Access
- Public
- Created
- Mar 2, 2022
- Updated
- Jul 20, 2021
- SHA
- 12c1152e
Try a prompt
Model card
From the Hugging Face model README
Model information
- language : Korean
- fine tuning data : kor_3i4k
- License : CC-BY-SA 4.0
- Base model : bert-base-multilingual-cased
- input : sentence
- output : intent
Train information
- train_runtime: 2376.638
- train_steps_per_second: 2.175
- train_loss: 0.356829648599977
- epoch: 3.0
How to use
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained (
"seongju/kor-3i4k-bert-base-cased"
)
model = AutoModelForSequenceClassification.from_pretrained (
"seongju/kor-3i4k-bert-base-cased"
)
inputs = tokenizer(
"너는 지금 무엇을 하고 있니?",
padding=True, truncation=True, max_length=128, return_tensors="pt"
)
outputs = model(**inputs)
probs = outputs[0].softmax(1)
output = probs.argmax().item()
- Task
- Text Classification
- Library
- transformers
- Type
- bert
- Created
- Mar 2, 2022
- Updated
- Jul 20, 2021