search

thinkingmachines / ph-poverty-mapping

Mapping Philippine Poverty using Machine Learning, Satellite Imagery, and Crowd-sourced Geospatial Information
https://stories.thinkingmachin.es/philippines-most-vulnerable-communities/
MIT License
80 stars 31 forks source link

boundaries CSV files #34

Closed sabman closed 3 years ago

sabman commented 3 years ago

Quick question. I am trying to reproduce the work and while running the 03_transfer_model.ipynb

It is expecting the following files which seem to be missing:

dhs_provinces_file = data_dir+'dhs_provinces.csv'
dhs_regions_file = data_dir+'dhs_regions.csv'

I don't see where they were created. Should they have been generated in the preparation? If so It could be that the original DHS data mentioned in the 00_dhs_prep.ipynb (PHHR70DT/PHHR70FL.DTA, PHHR70DT/PHHR70FL.DO) isn't available instead I was able to find these :

dhs_file = dhs_zip + 'PHHR71DT/PHHR71FL.DTA'
dhs_dict_file = dhs_zip + 'PHHR71DT/PHHR71FL.DO'

Any pointers would be greatly appreciated! Thanks again for publishing your work and merry christmas! 🎄 🎅

Skerre commented 3 years ago

Dear Sabman,

I am having the same issue. Did anyone answer you with a solution?

Best, Martin

ardieorden commented 3 years ago

Hi @sabman , apologies for the late reply!

The dhs_regions.csv file was uploaded a few weeks ago. You can check it out here: https://github.com/thinkingmachines/ph-poverty-mapping/blob/master/data/dhs_regions.csv.

dhs_provinces.csv and dhs_regions.csv were not generated in the Jupyter Notebooks and were actually created using QGIS.

I will just ask @ibtingzon if she still has a copy of dhs_provinces.csv so that she can upload it to the repo.

ardieorden commented 3 years ago

@sabman In the meantime, I think you can remove the following variables:

sabman commented 3 years ago

Thanks @ardieorden trying it out today appreciate your taking the time to get back to me. 🙇