Collaboration with TensorLayer about zhusuan HOT 5 CLOSED

Oh Sorry I missed the message.

Thank you for reaching out. And glad that you are interested in ZhuSuan.
Just have a look through TL and see my favorite Lassagne API :)
Do you have any plan in mind at this time?

from zhusuan.

It was more like a message to open the discussion and why not set up a conference on WeChat or a group chat ;)

Unfortunately I am the only one not speaking chinese, so I guess @zsdonghao will be glad to speak you if you'd like.

From my point of view, TL focuses on providing as many tools and best practices for tensorflow as possible. Being to aggregate various usecases is what makes TL really cool and practical to use on a day-to-day basis.

With zhusuan you have a point that neither TF or TL did cover, it would be nice to integrate your approach in the vast quantity of tools we already provide. Many directions are possible, some of which are:

• create a new lib merging the good things of both
• merging one library in the other one (will require some adaptation work).
• TL already provide LambdaLayer class, maybe we could reproduce this behaviour and expose zhusuan over this behaviour.

I mean so many different directions are possible, it really depends on where you wanna go and if you want to go there with us ;)

I sincerely believe that the more developers and maintainers we are on such a project the more changes we have to go for a wider user adoption.

As a backlink, I created an issue on TL repository in order to track the discussion here: tensorlayer/TensorLayer#688

from zhusuan.

@DEKHTIARJonathan thanks Jonathan. @thjashin hello, feel free to let us know if you want to have a chat via Wechat, Skype or others ~ : D

from zhusuan.

@DEKHTIARJonathan @zsdonghao That sounds great. Would love to chat. How about Wechat?

from zhusuan.

hi, please add me: zsdh123

from zhusuan.