search

Sparklewerk / hypermap

Tools for exploring NFT collections, permissively MIT licensed
https://sparklewerk.com
MIT License
0 stars 0 forks source link

hypermap_cryptopunks.ipynb: use PyTink #7

Open JohnTigue opened 2 years ago

JohnTigue commented 2 years ago

First hacking was in: hypermapping_cryptopunks.ipynb; second was hypermapping_bganpunks.ipynb. In the latter there is a bunch of clean code, unlike the first-take desperate hacking of the former. Keep evolving that, for now in the *.ipynbs.

And the step after this is to move the code into a package, see ManyHands/hypermap#2.

JohnTigue commented 2 years ago

Here's a current output of hypermapping_cryptopunks.ipynb:

hypermap_cryptopunks_553_highlighted

Notice the little horizontal lines atop each punk. Looks like a wee bug. Unacceptable. Could hunt it down, or this is v1 code and the hypermapping_cryptopunks.ipynb work is v2, so fold that back into this.

JohnTigue commented 2 years ago

Curiously, the same artifacts are not showing up in the 2D plot even though it's the same code:

    a_punk = og_punks.crop(get_cropbox(x))
JohnTigue commented 2 years ago

Actually, don't fold it into hypermap_cryptopunks.ipynb. Rather fold it into PyTink instead. I.e., get started on that and have the two (Punks and Bastards) notebooks use PyTink. We now have two collections hypermapped. There's an old chestnut that three use cases are needed before "reusable code" can realistically be called such. So, as we hypermap collection #3 (not sure who that is yet but it sure ain't cryptokitties or any other CPC stooge) build out PyTink. Having a Python package will be a serious sign that Sparklewerk is a tech stack web3 platform company.

JohnTigue commented 2 years ago

See also https://github.com/Sparklewerk/hypermap/issues/27

JohnTigue commented 2 years ago

Actually, PyTink and hypermap may be separate packages, with one dependent on the other. Not sure. See https://github.com/Sparklewerk/hypermap/issues/2