Nish

@Karim-Baig I think we don't need images for task 3. Task 3 uses RNN for character-level classification. I guess that's also one reason why the task doesn't work for more...

@Karim-Baig one possibility is to use some character-level embedding. In this case (task 3), it's already doing that using RNN. However, I presume that we could pre-train those layers in...

> @NISH1001 had mentioned about passing both the text and positions as embedding to a CNN in order to better localize the key-value pairs in documents . Is there any...

@inkarar The best way to approach training the cascadetabnet is to go through mmdetection framework where we treat each table as an object. So, once you are able to get...

Ya. I was trying to write. But got bored easily. It can be a seemingly-simple script (but might be complex too while parsing?) I guess. I think it can be...

Re: We could do something like `playx --update webdriver` which invokes the download script. So the script doesn't have to be be dependent on any modules within playx itself. Just...

Also, probably there's already an external library/package to do that?

Thanks a lot @deepjyoti30 . I should figure out to allocate time for playx too I guess. Haven't really touched the codebase for many months. Haha.

@athen66 @kootenpv So, I encountered this repo while randomly fiddling with tools. Apparently, the access_points binary gives correct `List[dict]` if ran as an external script (for instance copy the __init__.py...

I think one way for this to work is by describing how we can add the `access_points` module to `$PYTHONPATH` env. I will try to figure out a better way...