Stream: helpdesk (published)

Topic: Alternatives to FluxML

Notification Bot (Jul 13 2021 at 19:38):

This topic was moved here from #machine-learning > Alternatives to FluxML by Mason Protter

Notification Bot (Jul 13 2021 at 19:39):

This topic was moved here from #general > Alternatives to FluxML by Mason Protter

Brenhin Keller (Jul 13 2021 at 19:42):

Knet.jl and Avalon.jl both look like cool pacakges that I'd enjoy knowing how to use. I have hardly even used Flux.jl much myself though because I just don't to that much with neural nets

QuBit (Jul 13 2021 at 20:13):

Brian Chen

My statements drawn from the table found HERE

Brian Chen (Jul 13 2021 at 20:19):

Ah, that is talking about the integration between MLJ and Flux (MLJFlux.jl), and not Flux itself.

Brian Chen (Jul 13 2021 at 20:20):

Brenhin Keller said:

Knet.jl and Avalon.jl both look like cool pacakges that I'd enjoy knowing how to use. I have hardly even used Flux.jl much myself though because I just don't to that much with neural nets

Yes, they're both very compelling and deserve more recognition!

QuBit (Jul 13 2021 at 20:32):

Brian Chen

Do you know what the contributors to the original
code I referenced were going for?

I was attempting to reproduce this code in Pluto
succeeded but, was unable to:

1. Create a prediction model
2. Develop a training graph
3. Obtain a sensitivity Parameter
4. Predict the best performer

Perhaps those criteria do not quite fit the
'Housing' data set?

This is why I am pursuing the other options
starting with MLJ.

Thanks,

Notification Bot (Jul 20 2021 at 10:13):

QuBit has marked this topic as unresolved.

Notification Bot (Jul 20 2021 at 10:13):

QuBit has marked this topic as resolved.

Notification Bot (Jul 20 2021 at 10:14):

QuBit has marked this topic as unresolved.

Last updated: Oct 02 2023 at 04:34 UTC