Conversation
… test tasks, and tidy pixi dependencies
* main: (46 commits) Chage summary stats geoparquet filepaths from `output` to `intermediate` (#299) Update data downloads page (#300) Bump prefix-dev/setup-pixi from 0.9.1 to 0.9.2 in the actions group (#298) Update data download documentation (#293) migrate vector input datasets to unified ingestion and remove unused datasets (#297) Fix duplicate `avg_name` (#296) fix California and Tennessee region IDs in staging automatic deploy (#294) Add additional region IDs to QA PR automatic deploy (#292) create a unified infrastructure for ingesting and processing input datasets (#289) Combine county, tract and block PMTiles layers into a single regions.pmtiles layer (#291) Pyramid (#284) Use buffered slices to remove edge effects from neighborhood operations (#288) Bumps up RAM for `write-aggregated-region-analysis-files` job (#290) fix block dataset path construction in wind risk regional aggregation (#282) Adds a bbox struct for region pmtiles (#281) compute Dask-backed data before assert_equal/assert_all_close (#283) pipeline and configuration improvements (#279) Add cached valid_region_ids.json and use it in ChunkingConfig (#280) Combining wind-smeared data and Riley BP + smoothing (#278) update-docs: add first draft of all docs pages (#275) ...
There was a problem hiding this comment.
@orianac, i'm unable to run this notebook because it appears it references some local files that weren't added to this branch. are these files available in our S3 bucket? if so, can you point me to their location. if not, can you upload them to the S3 bucket?
tracts_dict = {}
for statistic in ['corr_low', 'low_bias', 'high_bias']:
tracts_dict[statistic] = gpd.read_file(f'{statistic}_tracts_{version}.shp', index_col=0)
andersy005
left a comment
There was a problem hiding this comment.
for consistency, can we centralize our inputs/outputs to use the carbonplan-ocr bucket instead of introducing yet another bucket?
for example, we're currently loading inputs and saving outputs to carbonplan-risks:
# Loading input
states = gpd.read_file('s3://carbonplan-risks/shapefiles/cb_2018_us_state_20m.zip')
# Saving output
buildings_in_census_tracts.to_parquet(
f's3://carbonplan-risks/shapefiles/buildings_tracts_{version}_geo.parquet'
)could we migrate these to use carbonplan-ocr instead?
|
Next steps:
|
|
@orianac / @katamartin, i updated the following notebooks with the new data (v0.12.0)
@orianac, i wasn't able to update this notebook because it uses some files that appear to be available on your local branch
i'm going to merge this PR as we discussed on Friday. please open follow up PRs to fix anything i may have missed |
Adding more analysis notebooks for the docs site.
@andersy005 could you move the following notebooks such that they can appear on the docs site?
We'll want to update these with the latest prod run so that the figures are representative of our data, so hold off on merging until we do that.
Also, please add any comments - very appreciated!