hari-ag00
hari-ag00
try removing the load_tokenizer=True
maybe you could try generating synthetic data and finetune the model and add post processing validators
try the same onnx conversion in notebook version once, https://github.com/urchade/GLiNER/blob/main/examples/convert_to_onnx.ipynb
and try to use earlier gliner version, latest gliner version has some bugs in onnx conversion
@jitesh0034 try using 0.2.3
@manickavela29, Not sure, which version are you using? facing any issue?
> 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...
Check my sample notebook, I have used my fine-tuned Gliner model and tried converting ONNX, it's working [https://colab.research.google.com/drive/1lwGSOAOo_hVs2K8cMkRks2MEJho7H18f?usp=sharing](url)