Visual Question Answering
Transformers
PyTorch
Safetensors
Korean
English
pix2struct
image-text-to-text
text2text-generation
Instructions to use nuua/ko-deplot with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use nuua/ko-deplot with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "visual-question-answering" 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("visual-question-answering", model="nuua/ko-deplot")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("nuua/ko-deplot") model = AutoModelForMultimodalLM.from_pretrained("nuua/ko-deplot", device_map="auto") - Notebooks
- Google Colab
- Kaggle
please, help
#3
by shrms - opened
Hello, wonderful developer Nuua!
I am a student studying machine learning.
Currently, we are trying to create a model that interprets chart images in Korean.
I need your help
Please help me
I wonder how you did data input, output structure setting, preprocessing, and normalization work during fine tuning.
If you don't mind, can I get the code you designed?
shrms changed discussion title from λμμ£ΌμΈμ λ©μμ΄ λμ μ¬λ¬λΆ! to help
shrms changed discussion title from help to help please
shrms changed discussion title from help please to please, help