William Nadolski

I'm also receiving the same issue when attempting to run python script from command line. Attempts to execute within Jupyter result in the kernel dying. I have however been successful...

Success! I was able to resolve the segmentation error by specifying which 'flavor' to use when reading the table (lattice vs. stream). As an example: tables = camelot.read_pdf(pdf, flavor='stream')

> > Just to make sure we got it right - v2/v3 models weights are under non-commercial license. Code is under apache?? > > yes, the non-commercial license is on...

One option is to first chunk your input text using the associated tokenizer then score each text chunk and collect the results. ``` from gliner import GLiNER from transformers import...

I am still having issues with attempts to convert to ONNX using the latest v0.2.16. ``` import os import onnx from gliner import GLiNER chunk_size = 384 gliner_model = GLiNER.from_pretrained("gliner-community/gliner_large-v2.5",...