Image-to-Text
Transformers
PyTorch
English
Korean
multilingual
veld
feature-extraction
vision, language
pretrained model
custom_code
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("KETI-AIR/veld-base", trust_remote_code=True, dtype="auto")Quick Links
veld base
Pretrained Vision Encoder Text Decoder Model in Korean and English. See Github for more details.
How to use
from transformers import AutoProcessor, AutoModel
processor = AutoProcessor.from_pretrained("KETI-AIR/veld-base", trust_remote_code=True)
model = AutoModel.from_pretrained("KETI-AIR/veld-base", trust_remote_code=True)
You can use AutoTokenizer and AutoFeatureExtractor instead AutoProcessor.
You don't need to pass trust_remote_code=True for AutoTokenizer and AutoFeatureExtractor
from transformers import AutoFeatureExtractor, AutoTokenizer, AutoModel
feature_extractor = AutoFeatureExtractor.from_pretrained("KETI-AIR/veld-base")
tokenizer = AutoTokenizer.from_pretrained("KETI-AIR/veld-base")
model = AutoModel.from_pretrained("KETI-AIR/veld-base", trust_remote_code=True)
- Downloads last month
- 13
# Use a pipeline as a high-level helper # Warning: Pipeline type "image-to-text" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("image-to-text", model="KETI-AIR/veld-base", trust_remote_code=True)