dinosg

these ARE census blocks being mapped to VTD's. However the VTD's (for Texas, from the MGGG state archive for 2010) got 'buffered' to avoid a point defect that prevented a...

you have a link for that repo? What I see at the general link https://dataverse.harvard.edu/file.xhtml?fileId=5007853&version=17.0 is stuff on 2020 election results but not obviously combining 2010 and 2020 demographics. Missing...

there also is their repo with "crosswalks" https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/T9VMJO but not updated w/ 2020 census pop data, just the block shapes and 2019 ACS data.

the shapefiles I used were at: https://github.com/mggg-states/TX-shapefiles

OK, I reduced the learning weight to 0.01, that helped and I get a fit. However using the dbn routines in addition to the nn training doesn't help any (the...

you can probably do that if you want but then have to make sure that size of the last batch is truncated so you don't run out of data and...