search

biocore / mds-approximations

Multidimensional scaling algorithms for microbiology-ecology datasets.
6 stars 7 forks source link

Python=3? #8

Closed mortonjt closed 9 years ago

mortonjt commented 9 years ago

I've been compiling scmds and nystrom from pycogent.

Now, pycogent is built on python2. In our first meeting, we decided that we wanted to make this Python3 compatible.

However, this means that we need to make all of the functions ported from pycogent python3 compatible. Do we really care if this package is python3 compatible? This is just a benchmarking suite. I'm not sure if it is time well spent if we make everything python3 compatible. Any thoughts @ElDeveloper @antgonza @josenavas ?

antgonza commented 9 years ago

IMOO: we don't care.

ElDeveloper commented 9 years ago

:+1: @antgonza.

On (Nov-03-15|11:53), Antonio Gonzalez wrote:

IMOO: we don't care.

Reply to this email directly or view it on GitHub: https://github.com/biocore/mds-approximations/issues/8#issuecomment-153469049

josenavas commented 9 years ago

The only advantage that I can see of having it already in Py3 is that if we want to later add some of these methods back to skbio they'll be already in Py3 - so we are agreeing on delaying this burden given that the number of methods that we are going to transfer to skbio is less than the ones that we are benchmarking.

Is this correct?

antgonza commented 9 years ago

Yes, we should only fully transfer those that are possible good solutions. All the rest will still live here.

ElDeveloper commented 9 years ago

Hmmm, that's true, though getting all methods to be py2/3 compatible might be more time-consuming at first than make sure 1 of them is compatible. Theory says that's the case, but not sure what will happen in practice.

On (Nov-03-15|13:26), Jose Navas wrote:

The only advantage that I can see of having it already in Py3 is that if we want to later add some of these methods back to skbio they'll be already in Py3 - so we are agreeing on delaying this burden given that the number of methods that we are going to transfer to skbio is less than the ones that we are benchmarking.

Is this correct?

Reply to this email directly or view it on GitHub: https://github.com/biocore/mds-approximations/issues/8#issuecomment-153494255

mortonjt commented 9 years ago

Ok. Let's shoot for python2 only then.

I seriously doubt most of these methods will be ported to skbio.

On Tue, Nov 3, 2015 at 1:28 PM, Yoshiki Vázquez Baeza < notifications@github.com> wrote:

Hmmm, that's true, though getting all methods to be py2/3 compatible might be more time-consuming at first than make sure 1 of them is compatible. Theory says that's the case, but not sure what will happen in practice.

On (Nov-03-15|13:26), Jose Navas wrote:

The only advantage that I can see of having it already in Py3 is that if we want to later add some of these methods back to skbio they'll be already in Py3 - so we are agreeing on delaying this burden given that the number of methods that we are going to transfer to skbio is less than the ones that we are benchmarking.

Is this correct?

Reply to this email directly or view it on GitHub:

https://github.com/biocore/mds-approximations/issues/8#issuecomment-153494255

— Reply to this email directly or view it on GitHub https://github.com/biocore/mds-approximations/issues/8#issuecomment-153495079 .

ElDeveloper commented 9 years ago

Sounds good!

On (Nov-03-15|14:04), mortonjt wrote:

Ok. Let's shoot for python2 only then.

I seriously doubt most of these methods will be ported to skbio.

On Tue, Nov 3, 2015 at 1:28 PM, Yoshiki Vázquez Baeza < notifications@github.com> wrote:

Hmmm, that's true, though getting all methods to be py2/3 compatible might be more time-consuming at first than make sure 1 of them is compatible. Theory says that's the case, but not sure what will happen in practice.

On (Nov-03-15|13:26), Jose Navas wrote:

The only advantage that I can see of having it already in Py3 is that if we want to later add some of these methods back to skbio they'll be already in Py3 - so we are agreeing on delaying this burden given that the number of methods that we are going to transfer to skbio is less than the ones that we are benchmarking.

Is this correct?

Reply to this email directly or view it on GitHub:

https://github.com/biocore/mds-approximations/issues/8#issuecomment-153494255

— Reply to this email directly or view it on GitHub https://github.com/biocore/mds-approximations/issues/8#issuecomment-153495079 .

Reply to this email directly or view it on GitHub: https://github.com/biocore/mds-approximations/issues/8#issuecomment-153503953

ElDeveloper commented 9 years ago

Closing this, as we've all agreed on a resolution.