Notebooks don't run

[ianhbell]

I have thus far found no happiness trying to get the notebooks to run. I am most interested in the notebook pypge_play.ipynb, which when I run it complains:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-3-484767f9bef2> in <module>()
     16     print_timing = True,
     17     log_details = True,
---> 18     fitness_func = FF.normalized_size_score
     19     )
     20 

AttributeError: 'module' object has no attribute 'normalized_size_score'

Have the notebooks fallen out of sync with the core code? I tried to build based on both master and develop branches and no luck.

[verdverm]

Sorry, notebooks are probably our of date, the best end-to-end is in with
the megarun script flow.

I probably have local commits pending, not sure how behind GH is. Will take
a look later tonight.

On Tue, Jun 7, 2016, 6:25 PM Ian Bell [email protected] wrote:

I have thus far found no happiness trying to get the notebooks to run. I
am most interested in the notebook pypge_play.ipynb, which when I run it
complains:

---

AttributeError Traceback (most recent call last)
in ()
16 print_timing = True,
17 log_details = True,
---> 18 fitness_func = FF.normalized_size_score
19 )
20

AttributeError: 'module' object has no attribute 'normalized_size_score'

Have the notebooks fallen out of sync with the core code? I tried to build
based on both master and develop branches and no luck.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#4, or mute the thread
https://github.com/notifications/unsubscribe/ABU4CBgDfH4tU6EqNrj39GSLH3wkYSOTks5qJgv8gaJpZM4IwfL1
.

[verdverm]

It's likely the notebooks are out of date.

I eventually had too much processing to do to want to keep it in Notebooks.

[ibell]

That's really a pity... Having working notebooks would be great as minimal working examples. Going through the hassle of firing up several docker images to run anything is a bit annoying. I think I might be able to get @embg to update the notebooks now that we have some code that runs.

[dnhkng]

I was hoping to use the notebook interface as well. ibell, have you got a working fork you could share?

[ibell]

We are currently waiting on reply from @verdverm , in their current form they do not run, and need major updates to run.

[verdverm]

A few points...

1. To get pypge running in notebooks, probably need to push a new version
   to PyPi. This will run like molasses in January.
2. Think about this like a Spark or Hadoop platform. The deployment of the
   system is larger than a single machine and has a Python interface.
3. My next efforts would be rewriting the core, then the search algorithm.
   As is, it does not scale well with input features.
4. I work now and am trying to level up there. It is unlikely I will have
   much time for this. Sorry about that.
   Cheers,
   Tony

On Mon, Jun 27, 2016, 7:42 AM Ian Bell [email protected] wrote:

We are currently waiting on reply from @verdverm
https://github.com/verdverm , in their current form they do not run,
and need major updates to run.

—
You are receiving this because you were mentioned.

Reply to this email directly, view it on GitHub
#4 (comment), or mute
the thread
https://github.com/notifications/unsubscribe/ABU4CBFnPXfqNUtsvCGqgv9xQ708zp_oks5qP9NTgaJpZM4IwfL1
.

[ibell]

@verdverm - It's a pity you don't have more time to dig into pypge stuff, but I totally understand. I've been developing another project in my free time, and I sympathize completely. It's very hard to find time to make progress on it, even if you want to.

We are/were super excited about the prospects of pypge, but our excitement has waned significantly. My student has the chops to make pypge really very useful, but he just needs a bit of assistance. I gather that assistance is probably not forthcoming. Again, a real tragedy. pypge has really such an exciting amount of potential.

P.S. I'll send you an email from work in the next day or so.