Instructions to use impira/layoutlm-document-qa with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use impira/layoutlm-document-qa with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("document-question-answering", model="impira/layoutlm-document-qa")# Load model directly from transformers import AutoTokenizer, AutoModelForDocumentQuestionAnswering tokenizer = AutoTokenizer.from_pretrained("impira/layoutlm-document-qa") model = AutoModelForDocumentQuestionAnswering.from_pretrained("impira/layoutlm-document-qa") - Notebooks
- Google Colab
- Kaggle
Image size invariance
#18
by acatovic - opened
For exactly the same image, I get very different results depending on what size (width x height) the image is set. Is this "invariance issue" a known problem, or is there some way to overcome or minimize it? Are there some guidelines around image size/resolution?
I have found the same issue. A higher resolution image is giving the best results.