Shannon Shen
Shannon Shen
# What does this PR do? Currently when we want to register a new config+tokenizer+model, per [the instructions](https://huggingface.co/docs/transformers/model_doc/auto), it seems we should do the following: ``` from transformers import AutoConfig,...
Even something like pixel diff should be helpful, given the bug reported in #131 ..
Example, ``` lp.AutoLayoutModel("lp://detectron1/PubLayNet") --> returns None ```
**Describe the bug** When using Python 3.9, it might fail to download the model files and eject the following errors: ``` ssl.SSLCertVerificationError: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get...
`box.crop_image(image)` doesn't support for PIL images. Current method requires manual conversion `box.crop_image(np.array(image))` beforehand.
**Describe the bug** When the input sequence is ordered differently from the element ids, the `lp.draw_box` will create inconsistent id annotation in the visualization. **To Reproduce** Example: ``` background =...
Substitute the previous hard-coded value `80` with a more robust version.