Thibault Clérice
Thibault Clérice
Hi there ! I found out that a GT section was added while I was tempted to create my own awesome list. One thing that I think would be great...
When running ketos-compile, I run into this specific error on multiple file. From eScriptorium viz or the XML, I can't catch what's wrong with this specific line. Files attached to...
The current multi-worker is broken on mail for `ketos compile`, I had to write a work-around for a script. The error thrown is `lxml.etree.ElementTree is unpickable` at the imap call....
Hey there :) I have been looking at the paper from Teklia which uses Doc-UFCN (https://gitlab.com/teklia/dla/doc-ufcn/-/blob/main/doc_ufcn/model.py) and mAP for evaluating line segmentation. I was wondering if, in the current state...
Hey :) Doing segmentation on a corpus, I had two self intersection errors. I don't know if this is due to a bad mask, a wrong geometry implementation or if...
Actually duplicates in part #53 because I forgot to go back to master... This adds the ability to chose on which metric to fit in regard to devset: accuracy, f1,...
…ched EOS for prediction (See #73) The current prediction time is quite slow, we agree that there might be room for improvement. After having a good look at it, it...
Hey ! I started reading about some other optimizers, as things went through my news feed (stuff like [this](https://penseeartificielle.fr/meilleur-optimizer-ranger-radam-lookahead-fastai/) or [that](https://medium.com/@lessw/new-deep-learning-optimizer-ranger-synergistic-combination-of-radam-lookahead-for-the-best-of-2dc83f79a48d)). I ended up trying to implement it in pie...
The current prediction time is quite slow, we agree that there might be room for improvement. After having a good look at it, it seemed clear that we were computing...
Akin to https://github.com/emanjavacas/pie/issues/76, I would like to try Dice Loss for Linear tasks. It's a bit more tricky to implement as it is done in model, and not in Trainer....