Nicolas Playe

I am inspecting the baudm code on Parseq (https://github.com/baudm/parseq). I think the code might not be too complicated to integrate. On my side I managed to connect your ocr_prediction and...

Hi, ok for timm. Other question: can we integrate 'pytorch-lightning~=1.6.5' to the requirements-pt.txt ?

ok, it sounds good for me :) I have added parseq class on my fork. Now I need to match all args of each methods between your wrapper and the...

@charlesmindee, what kind of data do you use to train your text recogntions models ? Do you use custom data ? and do you use special data for french documents...

Oh thanks it works for me now, I've modified something in the code. The second problem is when I want to evaluate the model with the metric STOI, I don't...

Yes it is ! thanks a lot !

Hello, I am currently implementing ParSeq. It is now working with quite good predictions :) Do you have any good advice to do a good pull request ? best

48 h running for the moment

Hello, I have tried multiple modifications on the code but couldn't succeed to do what I want. This is ths code of optimizer_overparametrization I have ```python class overparametrization_loss(nn.Module): def __init__(self,...

Hi ;) Tanks for your well documented reply ! I think by changing with another model in each iterations (resnet, vit, mobilenet, clip, etc.) can improve too ! I know...