[FRIAM] Google's TPUs

[Pieter Steenekamp]

I haven't used Google's TPU's. Why not? Because I prefer using H2O for my
deep learning applications and not Google's Tensorflow.

I evaluated Tensorflow and H2O and found H2O works better for me. I'm not
claiming a blanket "H2O is better than Tensorflow", it obviously depends on
many things.
And I also admit I did not do a very thorough evaluation,  but I found that
for me it's easier to get a "good enough" solution using H2O than with
Tensorflow. With my focus being more on the application as opposed to the
deep learning technology, H2O works better for me.
Further fact that I prefer R to Python, for similar reasons, also
contributed to my decision to use H2O rather than Tensorflow.
I did not evaluate other deep learning technologies.

But, I realize things are developing and changing very rapidly on all
fronts. TPU's could make me change my mind, but then, the "driverless AI"
of H2O could entice me to stay with H2O?

On 20 January 2018 at 18:23, Marcus Daniels <marcus at snoutfarm.com> wrote:

> Hi,
>
>
>
> Has anyone used Google’s compute cloud TPUs?
>
>
>
> Marcus
>
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20180120/f77d0bf4/attachment.html>