01-10-2020, 02:55 AM
(30-09-2020, 07:50 PM)Gioynich wrote: You are not allowed to view links. Register or Login to view.JKP, if you could share your full data set on the zodiacs (I know it was a lot of work, so I perfectly understand if you don't want to), I would be interested in building a machine learning classifier that, given some input features (e.g. is virgo male? is sagittarius human? etc.) predicts the most likely location and time period for the VM.
If you don't want to share all the images and you already have features for each zodiac you collected (including time and location), that would also work.
Let me know what you think
Yes, it was a lot of work. It took me about 8 or 9 years to find that many. Since 2017, I've only found a few new ones. The more you have, the harder it is to find more.
I didn't use the Warburg Institution collection (in fact I started the collection years before I knew about it and when they didn't have that much data), and I didn't find very many through Google Images either. I found 99% of them by going through manuscripts.
We already know the answer to the time period. It's expressed in the fashions, the style of the faces, and the thematic content. My blogs and Koen's blogs looked into that in some detail. Some of the maps I posted in those blogs include dates.