publicdata/j40-cejst-2
1
0
Fork 0
mirror of https://github.com/DOI-DO/j40-cejst-2.git synced 2025-07-26 16:41:16 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

Backend release branch to main (#1822)

Browse source 
* Create deploy_be_staging.yml (#1575)

* Imputing income using geographic neighbors (#1559)

Imputes income field with a light refactor. Needs more refactor and more tests (I spotchecked). Next ticket will check and address but a lot of "narwhal" architecture is here.

* Adding HOLC indicator (#1579)

Added HOLC indicator (Historic Redlining Score) from NCRC work; included 3.25 cutoff and low income as part of the housing burden category.

* Update backend for Puerto Rico (#1686)

* Update PR threshold count to 10

We now show 10 indicators for PR. See the discussion on the github issue for more info: https://github.com/usds/justice40-tool/issues/1621

* Do not use linguistic iso for Puerto Rico

Closes 1350.

Co-authored-by: Shelby Switzer <shelbyswitzer@gmail.com>

* updating

* Do not drop Guam and USVI from ETL (#1681)

* Remove code that drops Guam and USVI from ETL

* Add back code for dropping rows by FIPS code

We may want this functionality, so let's keep it and just make the constant currently be an empty array.

Co-authored-by: Shelby Switzer <shelbyswitzer@gmail.com>

* Emma nechamkin/holc patch (#1742)

Removing HOLC calculation from score narwhal.

* updating ejscreen data, try two (#1747)

* Rescaling linguistic isolation  (#1750)

Rescales linguistic isolation to drop puerto rico

* adds UST indicator (#1786)

adds leaky underground storage tanks

* Changing LHE in tiles to a boolean (#1767)

also includes merging / clean up of the release

* added indoor plumbing to chas

* added indoor plumbing to score housing burden

* added indoor plumbing to score housing burden

* first run through

* Refactor DOE Energy Burden and COI to use YAML (#1796)

* added tribalId for Supplemental dataset (#1804)

* Setting zoom levels for tribal map (#1810)

* NRI dataset and initial score YAML configuration (#1534)

* update be staging gha

* NRI dataset and initial score YAML configuration

* checkpoint

* adding data checks for release branch

* passing tests

* adding INPUT_EXTRACTED_FILE_NAME to base class

* lint

* columns to keep and tests

* update be staging gha

* checkpoint

* update be staging gha

* NRI dataset and initial score YAML configuration

* checkpoint

* adding data checks for release branch

* passing tests

* adding INPUT_EXTRACTED_FILE_NAME to base class

* lint

* columns to keep and tests

* checkpoint

* PR Review

* renoving source url

* tests

* stop execution of ETL if there's a YAML schema issue

* update be staging gha

* adding source url as class var again

* clean up

* force cache bust

* gha cache bust

* dynamically set score vars from YAML

* docsctrings

* removing last updated year - optional reverse percentile

* passing tests

* sort order

* column ordening

* PR review

* class level vars

* Updating DatasetsConfig

* fix pylint errors

* moving metadata hint back to code

Co-authored-by: lucasmbrown-usds <lucas.m.brown@omb.eop.gov>

* Correct copy typo (#1809)

* Add basic test suite for COI (#1518)

* Update COI to use new yaml (#1518)

* Add tests for DOE energy budren (1518

* Add dataset config for energy budren (1518)

* Refactor ETL to use datasets.yml (#1518)

* Add fake GEOIDs to COI tests (#1518)

* Refactor _setup_etl_instance_and_run_extract to base (#1518)

For the three classes we've done so far, a generic
_setup_etl_instance_and_run_extract will work fine, for the moment we
can reuse the same setup method until we decide future classes need more
flexibility --- but they can also always subclass so...

* Add output-path tests (#1518)

* Update YAML to match constant (#1518)

* Don't blindly set float format (#1518)

* Add defaults for extract (#1518)

* Run YAML load on all subclasses (#1518)

* Update description fields (#1518)

* Update YAML per final format (#1518)

* Update fixture tract IDs (#1518)

* Update base class refactor (#1518)

Now that NRI is final I needed to make a small number of updates to my
refactored code.

* Remove old comment (#1518)

* Fix type signature and return (#1518)

* Update per code review (#1518)

Co-authored-by: Jorge Escobar <83969469+esfoobar-usds@users.noreply.github.com>
Co-authored-by: lucasmbrown-usds <lucas.m.brown@omb.eop.gov>
Co-authored-by: Vim <86254807+vim-usds@users.noreply.github.com>

* Update etl_score_geo.py

Yikes! Fixing merge messup!

* Create deploy_be_staging.yml (#1575)

* Imputing income using geographic neighbors (#1559)

Imputes income field with a light refactor. Needs more refactor and more tests (I spotchecked). Next ticket will check and address but a lot of "narwhal" architecture is here.

* Adding HOLC indicator (#1579)

Added HOLC indicator (Historic Redlining Score) from NCRC work; included 3.25 cutoff and low income as part of the housing burden category.

* Update backend for Puerto Rico (#1686)

* Update PR threshold count to 10

We now show 10 indicators for PR. See the discussion on the github issue for more info: https://github.com/usds/justice40-tool/issues/1621

* Do not use linguistic iso for Puerto Rico

Closes 1350.

Co-authored-by: Shelby Switzer <shelbyswitzer@gmail.com>

* updating

* Do not drop Guam and USVI from ETL (#1681)

* Remove code that drops Guam and USVI from ETL

* Add back code for dropping rows by FIPS code

We may want this functionality, so let's keep it and just make the constant currently be an empty array.

Co-authored-by: Shelby Switzer <shelbyswitzer@gmail.com>

* Emma nechamkin/holc patch (#1742)

Removing HOLC calculation from score narwhal.

* updating ejscreen data, try two (#1747)

* Rescaling linguistic isolation  (#1750)

Rescales linguistic isolation to drop puerto rico

* adds UST indicator (#1786)

adds leaky underground storage tanks

* Changing LHE in tiles to a boolean (#1767)

also includes merging / clean up of the release

* added indoor plumbing to chas

* added indoor plumbing to score housing burden

* added indoor plumbing to score housing burden

* first run through

* Refactor DOE Energy Burden and COI to use YAML (#1796)

* added tribalId for Supplemental dataset (#1804)

* Setting zoom levels for tribal map (#1810)

* NRI dataset and initial score YAML configuration (#1534)

* update be staging gha

* NRI dataset and initial score YAML configuration

* checkpoint

* adding data checks for release branch

* passing tests

* adding INPUT_EXTRACTED_FILE_NAME to base class

* lint

* columns to keep and tests

* update be staging gha

* checkpoint

* update be staging gha

* NRI dataset and initial score YAML configuration

* checkpoint

* adding data checks for release branch

* passing tests

* adding INPUT_EXTRACTED_FILE_NAME to base class

* lint

* columns to keep and tests

* checkpoint

* PR Review

* renoving source url

* tests

* stop execution of ETL if there's a YAML schema issue

* update be staging gha

* adding source url as class var again

* clean up

* force cache bust

* gha cache bust

* dynamically set score vars from YAML

* docsctrings

* removing last updated year - optional reverse percentile

* passing tests

* sort order

* column ordening

* PR review

* class level vars

* Updating DatasetsConfig

* fix pylint errors

* moving metadata hint back to code

Co-authored-by: lucasmbrown-usds <lucas.m.brown@omb.eop.gov>

* Correct copy typo (#1809)

* Add basic test suite for COI (#1518)

* Update COI to use new yaml (#1518)

* Add tests for DOE energy budren (1518

* Add dataset config for energy budren (1518)

* Refactor ETL to use datasets.yml (#1518)

* Add fake GEOIDs to COI tests (#1518)

* Refactor _setup_etl_instance_and_run_extract to base (#1518)

For the three classes we've done so far, a generic
_setup_etl_instance_and_run_extract will work fine, for the moment we
can reuse the same setup method until we decide future classes need more
flexibility --- but they can also always subclass so...

* Add output-path tests (#1518)

* Update YAML to match constant (#1518)

* Don't blindly set float format (#1518)

* Add defaults for extract (#1518)

* Run YAML load on all subclasses (#1518)

* Update description fields (#1518)

* Update YAML per final format (#1518)

* Update fixture tract IDs (#1518)

* Update base class refactor (#1518)

Now that NRI is final I needed to make a small number of updates to my
refactored code.

* Remove old comment (#1518)

* Fix type signature and return (#1518)

* Update per code review (#1518)

Co-authored-by: Jorge Escobar <83969469+esfoobar-usds@users.noreply.github.com>
Co-authored-by: lucasmbrown-usds <lucas.m.brown@omb.eop.gov>
Co-authored-by: Vim <86254807+vim-usds@users.noreply.github.com>

* Update etl_score_geo.py

Yikes! Fixing merge messup!

* updated to fix linting errors (#1818)

Cleans and updates base branch

* Adding back MapComparison video

* Add FUDS ETL (#1817)

* Add spatial join method (#1871)

Since we'll need to figure out the tracts for a large number of points
in future tickets, add a utility to handle grabbing the tract geometries
and adding tract data to a point dataset.

* Add FUDS, also jupyter lab (#1871)

* Add YAML configs for FUDS (#1871)

* Allow input geoid to be optional (#1871)

* Add FUDS ETL, tests, test-datae noteobook (#1871)

This adds the ETL class for Formerly Used Defense Sites (FUDS). This is
different from most other ETLs since these FUDS are not provided by
tract, but instead by geographic point, so we need to assign FUDS to
tracts and then do calculations from there.

* Floats -> Ints, as I intended (#1871)

* Floats -> Ints, as I intended (#1871)

* Formatting fixes (#1871)

* Add test false positive GEOIDs (#1871)

* Add gdal binaries (#1871)

* Refactor pandas code to be more idiomatic (#1871)

Per Emma, the more pandas-y way of doing my counts is using np.where to
add the values i need, then groupby and size. It is definitely more
compact, and also I think more correct!

* Update configs per Emma suggestions (#1871)

* Type fixed! (#1871)

* Remove spurious import from vscode (#1871)

* Snapshot update after changing col name (#1871)

* Move up GDAL (#1871)

* Adjust geojson strategy (#1871)

* Try running census separately first (#1871)

* Fix import order (#1871)

* Cleanup cache strategy (#1871)

* Download census data from S3 instead of re-calculating (#1871)

* Clarify pandas code per Emma (#1871)

* Disable markdown check for link

* Adding DOT composite to travel score (#1820)

This adds the DOT dataset to the ETL and to the score. Note that currently we take a percentile of an average of percentiles.

* Adding first street foundation data (#1823)

Adding FSF flood and wildfire risk datasets to the score.

* first run -- adding NCLD data to the ETL, but not yet to the score

* Add abandoned mine lands data (#1824)

* Add notebook to generate test data (#1780)

* Add Abandoned Mine Land data (#1780)

Using a similar structure but simpler apporach compared to FUDs, add an
indicator for whether a tract has an abandonded mine.

* Adding some detail to dataset readmes

Just a thought!

* Apply feedback from revieiw (#1780)

* Fixup bad string that broke test (#1780)

* Update a string that I should have renamed (#1780)

* Reduce number of threads to reduce memory pressure (#1780)

* Try not running geo data (#1780)

* Run the high-memory sets separately (#1780)

* Actually deduplicate (#1780)

* Add flag for memory intensive ETLs (#1780)

* Document new flag for datasets (#1780)

* Add flag for new datasets fro rebase (#1780)

Co-authored-by: Emma Nechamkin <97977170+emma-nechamkin@users.noreply.github.com>

* Adding NLCD data (#1826)

Adding NLCD's natural space indicator end to end to the score.

* Add donut hole calculation to score (#1828)

Adds adjacency index to the pipeline. Requires thorough QA

* Adding eamlis and fuds data to legacy pollution in score (#1832)

Update to add EAMLIS and FUDS data to score

* Update to use new FSF files (#1838)

backend is partially done!

* Quick fix to kitchen or plumbing indicator

Yikes! I think I messed something up and dropped the pctile field suffix from when the KP score gets calculated. Fixing right quick.

* Fast flag update (#1844)

Added additional flags for the front end based on our conversation in stand up this morning.

* Tiles fix (#1845)

Fixes score-geo and adds flags

* Update etl_score_geo.py

* Issue 1827: Add demographics to tiles and download files (#1833)

* Adding demographics for use in sidebar and download files

* Updates backend constants to N (#1854)

* updated to show T/F/null vs T/F for AML and FUDS (#1866)

* fix markdown

* just testing that the boolean is preserved on gha

* checking drop tracts works

* OOPS!

Old changes persisted

* adding a check to the agvalue calculation for nri

* updated with error messages

* updated error message

* tuple type

* Score tests (#1847)

* update Python version on README; tuple typing fix

* Alaska tribal points fix (#1821)

* Bump mistune from 0.8.4 to 2.0.3 in /data/data-pipeline (#1777)

Bumps [mistune](https://github.com/lepture/mistune) from 0.8.4 to 2.0.3.
- [Release notes](https://github.com/lepture/mistune/releases)
- [Changelog](https://github.com/lepture/mistune/blob/master/docs/changes.rst)
- [Commits](https://github.com/lepture/mistune/compare/v0.8.4...v2.0.3)

---
updated-dependencies:
- dependency-name: mistune
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* poetry update

* initial pass of score tests

* add threshold tests

* added ses threshold (not donut, not island)

* testing suite -- stopping for the day

* added test for lead proxy indicator

* Refactor score tests to make them less verbose and more direct (#1865)

* Cleanup tests slightly before refactor (#1846)

* Refactor score calculations tests

* Feedback from review

* Refactor output tests like calculatoin tests (#1846) (#1870)

* Reorganize files (#1846)

* Switch from lru_cache to fixture scorpes (#1846)

* Add tests for all factors (#1846)

* Mark smoketests and run as part of be deply (#1846)

* Update renamed var (#1846)

* Switch from named tuple to dataclass (#1846)

This is annoying, but pylint in python3.8 was crashing parsing the named
tuple. We weren't using any namedtuple-specific features, so I made the
type a dataclass just to get pylint to behave.

* Add default timout to requests (#1846)

* Fix type (#1846)

* Fix merge mistake on poetry.lock (#1846)

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: Jorge Escobar <jorge.e.escobar@omb.eop.gov>
Co-authored-by: Jorge Escobar <83969469+esfoobar-usds@users.noreply.github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Matt Bowen <83967628+mattbowen-usds@users.noreply.github.com>
Co-authored-by: matt bowen <matthew.r.bowen@omb.eop.gov>

* just testing that the boolean is preserved on gha (#1867)

* updated with hopefully a fix; coercing aml, fuds, hrs to booleans for the raw value to preserve null character.

* Adding tests to ensure proper calculations (#1871)

* just testing that the boolean is preserved on gha
* checking drop tracts works
* adding a check to the agvalue calculation for nri
* updated with error messages

* tribal tiles fix (#1874)

* Alaska tribal points fix (#1821)

* tribal tiles fix

* disabling child opportunity

* lint

* removing COI

* removing commented out code

* Pipeline tile tests (#1864)

* temp update

* updating with fips check

* adding check on pfs

* updating with pfs test

* Update test_tiles_smoketests.py

* Fix lint errors (#1848)

* Add column names test (#1848)

* Mark tests as smoketests (#1848)

* Move to other score-related tests (#1848)

* Recast Total threshold criteria exceeded to int (#1848)

In writing tests to verify the output of the tiles csv matches the final
score CSV, I noticed TC/Total threshold criteria exceeded was getting
cast from an int64 to a float64 in the process of PostScoreETL. I
tracked it down to the line where we merge the score dataframe with
constants.DATA_CENSUS_CSV_FILE_PATH --- there where > 100 tracts in the
national census CSV that don't exist in the score, so those ended up
with a Total threshhold count of np.nan, which is a float, and thereby
cast those columns to float. For the moment I just cast it back.

* No need for low memeory (#1848)

* Add additional tests of tiles.csv (#1848)

* Drop pre-2010 rows before computing score (#1848)

Note this is probably NOT the optimal place for this change; it might
make more sense for each source to filter its own tracts down to the
acceptable tract list. However, that would be a pretty invasive change,
where this is central and plenty of other things are happening in score
transform that could be moved to sources, so for today, here's where the
change will live.

* Fix typo (#1848)

* Switch from filter to inner join (#1848)

* Remove no-op lines from tiles (#1848)

* Apply feedback from review, linter (#1848)

* Check the values oeverything in the frame (#1848)

* Refactor checker class (#1848)

* Add test for state names (#1848)

* cleanup from reviewing my own code (#1848)

* Fix lint error (#1858)

* Apply Emma's feedback from review (#1848)

* Remove refs to national_df (#1848)

* Account for new, fake nullable bools in tiles (#1848)

To handle a geojson limitation, Emma converted some nullable boolean
colunms to float64 in the tiles export with the values {0.0, 1.0, nan},
giving us the same expressiveness. Sadly, this broke my assumption that
all columns between the score and tiles csvs would have the same dtypes,
so I need to account for these new, fake bools in my test.

* Use equals instead of my worse version (#1848)

* Missed a spot where we called _create_score_data (#1848)

* Update per safety (#1848)

Co-authored-by: matt bowen <matthew.r.bowen@omb.eop.gov>

* Add tests to make sure each source makes it to the score correctly (#1878)

* Remove unused persistent poverty from score (#1835)

* Test a few datasets for overlap in the final score (#1835)

* Add remaining data sources (#1853)

* Apply code-review feedback (#1835)

* Rearrange a little for readabililty (#1835)

* Add tract test (#1835)

* Add test for score values (#1835)

* Check for unmatched source tracts (#1835)

* Cleanup numeric code to plaintext (#1835)

* Make import more obvious (#1835)

* Updating traffic barriers to include low pop threshold (#1889)

Changing the traffic barriers to only be included for places with recorded population

* Remove no land tracts from map (#1894)

remove from map

* Issue 1831: missing life expectancy data from Maine and Wisconsin (#1887)

* Fixing missing states and adding tests for states to all classes

* Removing low pop tracts from FEMA population loss (#1898)

dropping 0 population from FEMA

* 1831 Follow up (#1902)

This code causes no functional change to the code. It does two things:

1. Uses difference instead of - to improve code style for working with sets.

2. Removes the line EXPECTED_MISSING_STATES = ["02", "15"], which is now redundant because of the line I added (in a previous pull request) of ALASKA_AND_HAWAII_EXPECTED_IN_DATA = False.

* Add tests for all non-census sources (#1899)

* Refactor CDC life-expectancy (1554)

* Update to new tract list (#1554)

* Adjust for tests (#1848)

* Add tests for cdc_places (#1848)

* Add EJScreen tests (#1848)

* Add tests for HUD housing (#1848)

* Add tests for GeoCorr (#1848)

* Add persistent poverty tests (#1848)

* Update for sources without zips, for new validation (#1848)

* Update tests for new multi-CSV but (#1848)

Lucas updated the CDC life expectancy data to handle a bug where two
states are missing from the US Overall download. Since virtually none of
our other ETL classes download multiple CSVs directly like this, it
required a pretty invasive new mocking strategy.

* Add basic tests for nature deprived (#1848)

* Add wildfire tests (#1848)

* Add flood risk tests (#1848)

* Add DOT travel tests (#1848)

* Add historic redlining tests (#1848)

* Add tests for ME and WI (#1848)

* Update now that validation exists (#1848)

* Adjust for validation (#1848)

* Add health insurance back to cdc places (#1848)

Ooops

* Update tests with new field (#1848)

* Test for blank tract removal (#1848)

* Add tracts for clipping behavior

* Test clipping and zfill behavior (#1848)

* Fix bad test assumption (#1848)

* Simplify class, add test for tract padding (#1848)

* Fix percentage inversion, update tests (#1848)

Looking through the transformations, I noticed that we were subtracting
a percentage that is usually between 0-100 from 1 instead of 100, and so
were endind up with some surprising results. Confirmed with lucasmbrown-usds

* Add note about first street data (#1848)

* Issue 1900: Tribal overlap with Census tracts (#1903)

* working notebook

* updating notebook

* wip

* fixing broken tests

* adding tribal overlap files

* WIP

* WIP

* WIP, calculated count and names

* working

* partial cleanup

* partial cleanup

* updating field names

* fixing bug

* removing pyogrio

* removing unused imports

* updating test fixtures to be more realistic

* cleaning up notebook

* fixing black

* fixing flake8 errors

* adding tox instructions

* updating etl_score

* suppressing warning

* Use projected CRSes, ignore geom types (#1900)

I looked into this a bit, and in general the geometry type mismatch
changes very little about the calculation; we have a mix of
multipolygons and polygons. The fastest thing to do is just not keep
geom type; I did some runs with it set to both True and False, and
they're the same within 9 digits of precision. Logically we just want to
overlaps, regardless of how the actual geometries are encoded between
the frames, so we can in this case ignore the geom types and feel OKAY.

I also moved to projected CRSes, since we are actually trying to do area
calculations and so like, we should. Again, the change is small in
magnitude but logically more sound.

* Readd CDC dataset config (#1900)

* adding comments to fips code

* delete unnecessary loggers

Co-authored-by: matt bowen <matthew.r.bowen@omb.eop.gov>

* Improve score test documentation based on Lucas's feedback (#1835) (#1914)

* Better document base on Lucas's feedback (#1835)

* Fix typo (#1835)

* Add test to verify GEOJSON matches tiles (#1835)

* Remove NOOP line (#1835)

* Move GEOJSON generation up for new smoketest (#1835)

* Fixup code format (#1835)

* Update readme for new somketest (#1835)

* Cleanup source tests (#1912)

* Move test to base for broader coverage (#1848)

* Remove duplicate line (#1848)

* FUDS needed an extra mock (#1848)

* Add tribal count notebook (#1917) (#1919)

* Add tribal count notebook (#1917)

* test without caching

* added comment

Co-authored-by: lucasmbrown-usds <lucas.m.brown@omb.eop.gov>

* Add tribal overlap to downloads (#1907)

* Add tribal data to downloads (#1904)

* Update test pickle with current cols (#1904)

* Remove text of tribe names from GeoJSON (#1904)

* Update test data (#1904)

* Add tribal overlap to smoketests (#1904)

* Issue 1910: Do not impute income for 0 population tracts (#1918)

* should be working, has unnecessary loggers

* removing loggers and cleaning up

* updating ejscreen tests

* adding tests and responding to PR feedback

* fixing broken smoke test

* delete smoketest docs

* updating click

* updating click

* Bump just jupyterlab (#1930)

* Fixing link checker (#1929)

* Update deps safety says are vulnerable (#1937) (#1938)

Co-authored-by: matt bowen <matt@mattbowen.net>

* Add demos for island areas (#1932)

* Backfill population in island areas (#1882)

* Update smoketest to account for backfills (#1882)

As I wrote in the commend:
We backfill island areas with data from the 2010 census, so if THOSE tracts
have data beyond the data source, that's to be expected and is fine to pass.
If some other state or territory does though, this should fail

This ends up being a nice way of documenting that behavior i guess!

* Fixup lint issues (#1882)

* Add in race demos to 2010 census pull (#1851)

* Add backfill data to score (#1851)

* Change column name (#1851)

* Fill demos after the score (#1851)

* Add income back, adjust test (#1882)

* Apply code-review feedback (#1851)

* Add test for island area backfill (#1851)

* Fix bad rename (#1851)

* Reorder download fields, add plumbing back (#1942)

* Add back lack of plumbing fields (#1920)

* Reorder fields for excel (#1921)

* Reorder excel fields (#1921)

* Fix formating, lint errors, pickes (#1921)

* Add missing plumbing col, fix order again (#1921)

* Update that pickle (#1921)

* refactoring tribal (#1960)

* updated with scoring comparison

* updated for narhwal -- leaving commented code in for now

* pydantic upgrade

* produce a string for the front end to ingest (#1963)

* wip

* i believe this works -- let's see the pipeline

* updated fixtures

* Adding ADJLI_ET (#1976)

* updated tile data

* ensuring adjli_et in

* Add back income percentile (#1977)

* Add missing field to download (#1964)

* Remove pydantic since it's unused (#1964)

* Add percentile to CSV (#1964)

* Update downloadable pickle (#1964)

* Issue 105: Configure and run `black` and other pre-commit hooks (clean branch) (#1962)

* Configure and run `black` and other pre-commit hooks

Co-authored-by: matt bowen <matthew.r.bowen@omb.eop.gov>

* Removing fixed python version for black (#1985)

* Fixup TA_COUNT and TA_PERC (#1991)

* Change TA_PERC, change TA_COUNT (#1988, #1989)

- Make TA_PERC_STR back into a nullable float following the rules
  requestsed in #1989
- Move TA_COUNT to be TA_COUNT_AK, also add a null TA_COUNT_C for CONUS
  that we can fill in later.

* Fix typo comment (#1988)

* Issue 1992: Do not impute income for null population tracts (#1993)

* Hotfix for DOT data source DNS issue (#1999)

* Make tribal overlap set score N (#2004)

* Add "Is a Tribal DAC" field (#1998)

* Add tribal DACs to score N final (#1998)

* Add new fields to downloads (#1998)

* Make a int a float (#1998)

* Update field names, apply feedback (#1998)

* Add assertions around codebook (#2014)

* Add assertion around codebook (#1505)

* Assert csv and excel have same cols (#1505)

* Remove suffixes from tribal lands (#1974) (#2008)

* Data source location (#2015)

* data source location

* toml

* cdc_places

* cdc_svi_index

* url updates

* child oppy and dot travel

* up to hud_recap

* completed ticket

* cache bust

* hud_recap

* us_army_fuds

* Remove vars the frontend doesn't use (#2020) (#2022)

I did a pretty rough and simple analysis of the variables we put in the
tiles and grepped the frontend code to see if (1) they're ever accessed
and (2) if they're used, even if they're read once. I removed everything
I noticed was not accessed.

* Disable file size limits on tiles (#2031)

* Disable file size limits on tiles

* Remove print debugs

I know.

* Update file name pattern (#2037) (#2038)

* Update file name pattern (#2037)

* Remove ETL from generation (2037)

I looked more carefully, and this ETL step isn't used in the score, so
there's no need to run it every time. Per previous steps, I removed it
from constants so the code is there it won't run by default.

* Round ALL the float fields for the tiles (#2040)

* Round ALL the float fields for the tiles (#2033)

* Floor in a simpler way (#2033)

Emma pointed out that all teh stuff we're doing in floor_series is
probably unnecessary for this case, so just use the built-in floor.

* Update pickle I missed (#2033)

* Clean commit of just aggregate burden notebook (#1819)

added a burden notebook

* Update the dockerfile (#2045)

* Update so the image builds (#2026)

* Fix bad dict (2026)

* Rename census tract field in downloads (#2068)

* Change tract ID field name (2060)

* Update lockfile (#2061)

* Bump safety, jupyter, wheel (#2061)

* DOn't depend directly on wheel (2061)

* Bring narwhal reqs in line with main

* Update tribal area counts (#2071)

* Rename tribal area field (2062)

* Add missing file (#2062)

* Add checks to create version (#2047) (#2052)

* Fix failing safety (#2114)

* Ignore vuln that doesn't affect us 2113

https://nvd.nist.gov/vuln/detail/CVE-2022-42969 landed recently and
there's no fix in py (which is maintenance mode). From my analysis, that
CVE cannot hurt us (famous last words), so we'll ignore the vuln for
now.

* 2113 Update our gdal ppa

* that didn't work (2113)

* Don't add the PPA, the package exists (#2113)

* Fix type (#2113)

* Force an update of wheel 2113

* Also remove PPA line from create-score-versions

* Drop 3.8 because of wheel 2113

* Put back 3.8, use newer actions

* Try another way of upgrading wheel 2113

* Upgrade wheel in tox too 2113

* Typo fix 2113

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: Emma Nechamkin <97977170+emma-nechamkin@users.noreply.github.com>
Co-authored-by: Shelby Switzer <shelby.c.switzer@omb.eop.gov>
Co-authored-by: Shelby Switzer <shelbyswitzer@gmail.com>
Co-authored-by: Emma Nechamkin <Emma.J.Nechamkin@omb.eop.gov>
Co-authored-by: Matt Bowen <83967628+mattbowen-usds@users.noreply.github.com>
Co-authored-by: Jorge Escobar <83969469+esfoobar-usds@users.noreply.github.com>
Co-authored-by: lucasmbrown-usds <lucas.m.brown@omb.eop.gov>
Co-authored-by: Jorge Escobar <jorge.e.escobar@omb.eop.gov>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: matt bowen <matthew.r.bowen@omb.eop.gov>
Co-authored-by: matt bowen <matt@mattbowen.net>
This commit is contained in:
Vim 2022-12-01 18:50:54 -08:00 • committed by GitHub
commit b97e60bfbb
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
285 changed files with 20485 additions and 3880 deletions
Download patch file Download diff file Expand all files Collapse all files

99
data/data-pipeline/data_pipeline/etl/base.py
View file

@ -5,15 +5,15 @@ import typing
from typing import Optional
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.score.schemas.datasets import DatasetsConfig
from data_pipeline.utils import (
load_yaml_dict_from_file,
unzip_file_from_url,
remove_all_from_dir,
get_module_logger,
from data_pipeline.etl.score.etl_utils import (
compare_to_list_of_expected_state_fips_codes,
)
from data_pipeline.etl.score.schemas.datasets import DatasetsConfig
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import load_yaml_dict_from_file
from data_pipeline.utils import remove_all_from_dir
from data_pipeline.utils import unzip_file_from_url
logger = get_module_logger(__name__)
@ -43,10 +43,11 @@ class ExtractTransformLoad:
APP_ROOT: pathlib.Path = settings.APP_ROOT
# Directories
DATA_PATH: pathlib.Path = APP_ROOT / "data"
DATA_PATH: pathlib.Path = settings.DATA_PATH
TMP_PATH: pathlib.Path = DATA_PATH / "tmp"
CONTENT_CONFIG: pathlib.Path = APP_ROOT / "content" / "config"
DATASET_CONFIG: pathlib.Path = APP_ROOT / "etl" / "score" / "config"
DATASET_CONFIG_PATH: pathlib.Path = APP_ROOT / "etl" / "score" / "config"
DATASET_CONFIG: Optional[dict] = None
# Parameters
GEOID_FIELD_NAME: str = "GEOID10"
@ -81,6 +82,23 @@ class ExtractTransformLoad:
# NULL_REPRESENTATION is how nulls are represented on the input field
NULL_REPRESENTATION: str = None
# Whether this ETL contains data for the continental nation (DC & the US states
# except for Alaska and Hawaii)
CONTINENTAL_US_EXPECTED_IN_DATA: bool = True
# Whether this ETL contains data for Alaska and Hawaii
ALASKA_AND_HAWAII_EXPECTED_IN_DATA: bool = True
# Whether this ETL contains data for Puerto Rico
PUERTO_RICO_EXPECTED_IN_DATA: bool = True
# Whether this ETL contains data for the island areas
ISLAND_AREAS_EXPECTED_IN_DATA: bool = False
# Whether this ETL contains known missing data for any additional
# states/territories
EXPECTED_MISSING_STATES: typing.List[str] = []
# Thirteen digits in a census block group ID.
EXPECTED_CENSUS_BLOCK_GROUPS_CHARACTER_LENGTH: int = 13
# TODO: investigate. Census says there are only 217,740 CBGs in the US. This might
@ -94,17 +112,24 @@ class ExtractTransformLoad:
# periods. https://github.com/usds/justice40-tool/issues/964
EXPECTED_MAX_CENSUS_TRACTS: int = 74160
# Should this dataset load its configuration from
# the YAML files?
LOAD_YAML_CONFIG: bool = False
# We use output_df as the final dataframe to use to write to the CSV
# It is used on the "load" base class method
output_df: pd.DataFrame = None
def __init_subclass__(cls) -> None:
if cls.LOAD_YAML_CONFIG:
cls.DATASET_CONFIG = cls.yaml_config_load()
@classmethod
def yaml_config_load(cls) -> dict:
"""Generate config dictionary and set instance variables from YAML dataset."""
# check if the class instance has score YAML definitions
datasets_config = load_yaml_dict_from_file(
cls.DATASET_CONFIG / "datasets.yml",
cls.DATASET_CONFIG_PATH / "datasets.yml",
DatasetsConfig,
)
@ -123,9 +148,10 @@ class ExtractTransformLoad:
sys.exit()
# set some of the basic fields
cls.INPUT_GEOID_TRACT_FIELD_NAME = dataset_config[
"input_geoid_tract_field_name"
]
if "input_geoid_tract_field_name" in dataset_config:
cls.INPUT_GEOID_TRACT_FIELD_NAME = dataset_config[
"input_geoid_tract_field_name"
]
# get the columns to write on the CSV
# and set the constants
@ -134,11 +160,7 @@ class ExtractTransformLoad:
]
for field in dataset_config["load_fields"]:
cls.COLUMNS_TO_KEEP.append(field["long_name"])
# set the constants for the class
setattr(cls, field["df_field_name"], field["long_name"])
# return the config dict
return dataset_config
# This is a classmethod so it can be used by `get_data_frame` without
@ -176,14 +198,18 @@ class ExtractTransformLoad:
to get the file from a source url, unzips it and stores it on an
extract_path."""
# this can be accessed via super().extract()
if source_url and extract_path:
unzip_file_from_url(
file_url=source_url,
download_path=self.get_tmp_path(),
unzipped_file_path=extract_path,
verify=verify,
)
if source_url is None:
source_url = self.SOURCE_URL
if extract_path is None:
extract_path = self.get_tmp_path()
unzip_file_from_url(
file_url=source_url,
download_path=self.get_tmp_path(),
unzipped_file_path=extract_path,
verify=verify,
)
def transform(self) -> None:
"""Transform the data extracted into a format that can be consumed by the
@ -280,6 +306,24 @@ class ExtractTransformLoad:
f"`{geo_field}`."
)
# Check whether data contains expected states
states_in_output_df = (
self.output_df[self.GEOID_TRACT_FIELD_NAME]
.str[0:2]
.unique()
.tolist()
)
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=states_in_output_df,
continental_us_expected=self.CONTINENTAL_US_EXPECTED_IN_DATA,
alaska_and_hawaii_expected=self.ALASKA_AND_HAWAII_EXPECTED_IN_DATA,
puerto_rico_expected=self.PUERTO_RICO_EXPECTED_IN_DATA,
island_areas_expected=self.ISLAND_AREAS_EXPECTED_IN_DATA,
additional_fips_codes_not_expected=self.EXPECTED_MISSING_STATES,
dataset_name=self.NAME,
)
def load(self, float_format=None) -> None:
"""Saves the transformed data.
@ -318,6 +362,9 @@ class ExtractTransformLoad:
f"No file found at `{output_file_path}`."
)
logger.info(
f"Reading in CSV `{output_file_path}` for ETL of class `{cls}`."
)
output_df = pd.read_csv(
output_file_path,
dtype={

99
data/data-pipeline/data_pipeline/etl/constants.py
View file

@ -3,121 +3,188 @@ DATASET_LIST = [
"name": "cdc_places",
"module_dir": "cdc_places",
"class_name": "CDCPlacesETL",
"is_memory_intensive": False,
},
{
"name": "national_risk_index",
"module_dir": "national_risk_index",
"class_name": "NationalRiskIndexETL",
"is_memory_intensive": False,
},
{
"name": "travel_composite",
"module_dir": "dot_travel_composite",
"class_name": "TravelCompositeETL",
"is_memory_intensive": False,
},
{
"name": "tree_equity_score",
"module_dir": "tree_equity_score",
"class_name": "TreeEquityScoreETL",
},
{
"name": "census_acs",
"module_dir": "census_acs",
"class_name": "CensusACSETL",
},
{
"name": "census_acs_2010",
"module_dir": "census_acs_2010",
"class_name": "CensusACS2010ETL",
"is_memory_intensive": False,
},
{
"name": "census_decennial",
"module_dir": "census_decennial",
"class_name": "CensusDecennialETL",
"is_memory_intensive": False,
},
{
"name": "mapping_for_ej",
"module_dir": "mapping_for_ej",
"class_name": "MappingForEJETL",
"is_memory_intensive": False,
},
{
"name": "fsf_flood_risk",
"module_dir": "fsf_flood_risk",
"class_name": "FloodRiskETL",
"is_memory_intensive": False,
},
{
"name": "fsf_wildfire_risk",
"module_dir": "fsf_wildfire_risk",
"class_name": "WildfireRiskETL",
"is_memory_intensive": False,
},
{
"name": "ejscreen",
"module_dir": "ejscreen",
"class_name": "EJSCREENETL",
"is_memory_intensive": False,
},
{
"name": "hud_housing",
"module_dir": "hud_housing",
"class_name": "HudHousingETL",
"is_memory_intensive": False,
},
{
"name": "nlcd_nature_deprived",
"module_dir": "nlcd_nature_deprived",
"class_name": "NatureDeprivedETL",
"is_memory_intensive": False,
},
{
"name": "census_acs_median_income",
"module_dir": "census_acs_median_income",
"class_name": "CensusACSMedianIncomeETL",
"is_memory_intensive": False,
},
{
"name": "cdc_life_expectancy",
"module_dir": "cdc_life_expectancy",
"class_name": "CDCLifeExpectancy",
"is_memory_intensive": False,
},
{
"name": "doe_energy_burden",
"module_dir": "doe_energy_burden",
"class_name": "DOEEnergyBurden",
"is_memory_intensive": False,
},
{
"name": "geocorr",
"module_dir": "geocorr",
"class_name": "GeoCorrETL",
},
{
"name": "child_opportunity_index",
"module_dir": "child_opportunity_index",
"class_name": "ChildOpportunityIndex",
"is_memory_intensive": False,
},
{
"name": "mapping_inequality",
"module_dir": "mapping_inequality",
"class_name": "MappingInequalityETL",
"is_memory_intensive": False,
},
{
"name": "persistent_poverty",
"module_dir": "persistent_poverty",
"class_name": "PersistentPovertyETL",
"is_memory_intensive": False,
},
{
"name": "ejscreen_areas_of_concern",
"module_dir": "ejscreen_areas_of_concern",
"class_name": "EJSCREENAreasOfConcernETL",
"is_memory_intensive": False,
},
{
"name": "calenviroscreen",
"module_dir": "calenviroscreen",
"class_name": "CalEnviroScreenETL",
"is_memory_intensive": False,
},
{
"name": "hud_recap",
"module_dir": "hud_recap",
"class_name": "HudRecapETL",
"is_memory_intensive": False,
},
{
"name": "epa_rsei",
"module_dir": "epa_rsei",
"class_name": "EPARiskScreeningEnvironmentalIndicatorsETL",
"is_memory_intensive": False,
},
{
"name": "energy_definition_alternative_draft",
"module_dir": "energy_definition_alternative_draft",
"class_name": "EnergyDefinitionAlternativeDraft",
"is_memory_intensive": False,
},
{
"name": "michigan_ejscreen",
"module_dir": "michigan_ejscreen",
"class_name": "MichiganEnviroScreenETL",
"is_memory_intensive": False,
},
{
"name": "cdc_svi_index",
"module_dir": "cdc_svi_index",
"class_name": "CDCSVIIndex",
"is_memory_intensive": False,
},
{
"name": "maryland_ejscreen",
"module_dir": "maryland_ejscreen",
"class_name": "MarylandEJScreenETL",
"is_memory_intensive": False,
},
{
"name": "historic_redlining",
"module_dir": "historic_redlining",
"class_name": "HistoricRedliningETL",
"is_memory_intensive": False,
},
# This has to come after us.json exists
{
"name": "census_acs",
"module_dir": "census_acs",
"class_name": "CensusACSETL",
"is_memory_intensive": False,
},
{
"name": "census_acs_2010",
"module_dir": "census_acs_2010",
"class_name": "CensusACS2010ETL",
"is_memory_intensive": False,
},
{
"name": "us_army_fuds",
"module_dir": "us_army_fuds",
"class_name": "USArmyFUDS",
"is_memory_intensive": True,
},
{
"name": "eamlis",
"module_dir": "eamlis",
"class_name": "AbandonedMineETL",
"is_memory_intensive": True,
},
{
"name": "tribal_overlap",
"module_dir": "tribal_overlap",
"class_name": "TribalOverlapETL",
"is_memory_intensive": True,
},
]
@ -125,10 +192,12 @@ CENSUS_INFO = {
"name": "census",
"module_dir": "census",
"class_name": "CensusETL",
"is_memory_intensive": False,
}
TRIBAL_INFO = {
"name": "tribal",
"module_dir": "tribal",
"class_name": "TribalETL",
"is_memory_intensive": False,
}

45
data/data-pipeline/data_pipeline/etl/runner.py
View file

@ -1,5 +1,5 @@
import importlib
import concurrent.futures
import importlib
import typing
from data_pipeline.etl.score.etl_score import ScoreETL
@ -77,16 +77,41 @@ def etl_runner(dataset_to_run: str = None) -> None:
"""
dataset_list = _get_datasets_to_run(dataset_to_run)
with concurrent.futures.ThreadPoolExecutor() as executor:
futures = {
executor.submit(_run_one_dataset, dataset=dataset)
for dataset in dataset_list
}
# Because we are memory constrained on our infrastructure,
# we split datasets into those that are not memory intensive
# (is_memory_intensive == False) and thereby can be safely
# run in parallel, and those that require more RAM and thus
# should be run sequentially. The is_memory_intensive_flag is
# set manually in constants.py based on experience running
# the pipeline
concurrent_datasets = [
dataset
for dataset in dataset_list
if not dataset["is_memory_intensive"]
]
high_memory_datasets = [
dataset for dataset in dataset_list if dataset["is_memory_intensive"]
]
for fut in concurrent.futures.as_completed(futures):
# Calling result will raise an exception if one occurred.
# Otherwise, the exceptions are silently ignored.
fut.result()
if concurrent_datasets:
logger.info("Running concurrent jobs")
with concurrent.futures.ThreadPoolExecutor() as executor:
futures = {
executor.submit(_run_one_dataset, dataset=dataset)
for dataset in concurrent_datasets
}
for fut in concurrent.futures.as_completed(futures):
# Calling result will raise an exception if one occurred.
# Otherwise, the exceptions are silently ignored.
fut.result()
# Note: these high-memory datasets also usually require the Census geojson to be
# generated, and one of them requires the Tribal geojson to be generated.
if high_memory_datasets:
logger.info("Running high-memory jobs")
for dataset in high_memory_datasets:
_run_one_dataset(dataset=dataset)
def score_generate() -> None:

258
data/data-pipeline/data_pipeline/etl/score/config/datasets.yml
View file

@ -35,7 +35,6 @@ datasets:
include_in_tiles: true
include_in_downloadable_files: true
create_percentile: true
- short_name: "ex_ag_loss"
df_field_name: "EXPECTED_AGRICULTURE_LOSS_RATE_FIELD_NAME"
long_name: "Expected agricultural loss rate (Natural Hazards Risk Index)"
@ -54,7 +53,6 @@ datasets:
include_in_tiles: true
include_in_downloadable_files: true
create_percentile: true
- short_name: "ex_bldg_loss"
df_field_name: "EXPECTED_BUILDING_LOSS_RATE_FIELD_NAME"
long_name: "Expected building loss rate (Natural Hazards Risk Index)"
@ -72,8 +70,262 @@ datasets:
include_in_tiles: true
include_in_downloadable_files: true
create_percentile: true
- short_name: "has_ag_val"
df_field_name: "CONTAINS_AGRIVALUE"
long_name: "Contains agricultural value"
field_type: bool
- long_name: "Child Opportunity Index 2.0 database"
short_name: "coi"
module_name: "child_opportunity_index"
input_geoid_tract_field_name: "geoid"
load_fields:
- short_name: "he_heat"
df_field_name: "EXTREME_HEAT_FIELD"
long_name: "Summer days above 90F"
field_type: float
include_in_downloadable_files: true
include_in_tiles: true
- short_name: "he_food"
long_name: "Percent low access to healthy food"
df_field_name: "HEALTHY_FOOD_FIELD"
field_type: float
include_in_downloadable_files: true
include_in_tiles: true
- short_name: "he_green"
long_name: "Percent impenetrable surface areas"
df_field_name: "IMPENETRABLE_SURFACES_FIELD"
field_type: float
include_in_downloadable_files: true
include_in_tiles: true
- short_name: "ed_reading"
df_field_name: "READING_FIELD"
long_name: "Third grade reading proficiency"
field_type: float
include_in_downloadable_files: true
include_in_tiles: true
- long_name: "Low-Income Energy Affordabililty Data"
short_name: "LEAD"
module_name: "doe_energy_burden"
input_geoid_tract_field_name: "FIP"
load_fields:
- short_name: "EBP_PFS"
df_field_name: "REVISED_ENERGY_BURDEN_FIELD_NAME"
long_name: "Energy burden"
field_type: float
include_in_downloadable_files: true
include_in_tiles: true
- long_name: "Formerly Used Defense Sites"
short_name: "FUDS"
module_name: "us_army_fuds"
load_fields:
- short_name: "fuds_count"
df_field_name: "ELIGIBLE_FUDS_COUNT_FIELD_NAME"
long_name: "Count of eligible Formerly Used Defense Site (FUDS) properties centroids"
description_short:
"The number of FUDS marked as Eligible and Has Project in the tract."
field_type: int64
include_in_tiles: false
include_in_downloadable_files: false
- short_name: "not_fuds_ct"
df_field_name: "INELIGIBLE_FUDS_COUNT_FIELD_NAME"
long_name: "Count of ineligible Formerly Used Defense Site (FUDS) properties centroids"
description_short:
"The number of FUDS marked as Ineligible or Project in the tract."
field_type: int64
include_in_tiles: false
include_in_downloadable_files: false
- short_name: "has_fuds"
df_field_name: "ELIGIBLE_FUDS_BINARY_FIELD_NAME"
long_name: "Is there at least one Formerly Used Defense Site (FUDS) in the tract?"
description_short:
"Whether the tract has a FUDS"
field_type: bool
include_in_tiles: false
include_in_downloadable_files: false
- long_name: "Abandoned Mine Land Inventory System"
short_name: "eAMLIS"
module_name: "eamlis"
load_fields:
- short_name: "has_aml"
df_field_name: "AML_BOOLEAN"
long_name: "Is there at least one abandoned mine in this census tract?"
description_short:
"Whether the tract has an abandoned mine"
field_type: bool
include_in_tiles: true
include_in_downloadable_files: true
- long_name: "Example ETL"
short_name: "Example"
module_name: "example_dataset"
input_geoid_tract_field_name: "GEOID10_TRACT"
load_fields:
- short_name: "EXAMPLE_FIELD"
df_field_name: "Input Field 1"
long_name: "Example Field 1"
field_type: float
include_in_tiles: true
include_in_downloadable_files: true
- long_name: "First Street Foundation Flood Risk"
short_name: "FSF Flood Risk"
module_name: fsf_flood_risk
input_geoid_tract_field_name: "GEOID"
load_fields:
- short_name: "flood_eligible_properties"
df_field_name: "COUNT_PROPERTIES"
long_name: "Count of properties eligible for flood risk calculation within tract (floor of 250)"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: false
- short_name: "flood_risk_properties_today"
df_field_name: "PROPERTIES_AT_RISK_FROM_FLOODING_TODAY"
long_name: "Count of properties at risk of flood today"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: false
- short_name: "flood_risk_properties_30yrs"
df_field_name: "PROPERTIES_AT_RISK_FROM_FLOODING_IN_30_YEARS"
long_name: "Count of properties at risk of flood in 30 years"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: false
- short_name: "flood_risk_share_today"
df_field_name: "SHARE_OF_PROPERTIES_AT_RISK_FROM_FLOODING_TODAY"
long_name: "Share of properties at risk of flood today"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: true
- short_name: "flood_risk_share_30yrs"
df_field_name: "SHARE_OF_PROPERTIES_AT_RISK_FROM_FLOODING_IN_30_YEARS"
long_name: "Share of properties at risk of flood in 30 years"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: true
- long_name: "First Street Foundation Wildfire Risk"
short_name: "FSF Wildfire Risk"
module_name: fsf_wildfire_risk
input_geoid_tract_field_name: "GEOID"
load_fields:
- short_name: "fire_eligible_properties"
df_field_name: "COUNT_PROPERTIES"
long_name: "Count of properties eligible for wildfire risk calculation within tract (floor of 250)"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: false
- short_name: "fire_risk_properties_today"
df_field_name: "PROPERTIES_AT_RISK_FROM_FIRE_TODAY"
long_name: "Count of properties at risk of wildfire today"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: false
- short_name: "fire_risk_properties_30yrs"
df_field_name: "PROPERTIES_AT_RISK_FROM_FIRE_IN_30_YEARS"
long_name: "Count of properties at risk of wildfire in 30 years"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: false
- short_name: "fire_risk_share_today"
df_field_name: "SHARE_OF_PROPERTIES_AT_RISK_FROM_FIRE_TODAY"
long_name: "Share of properties at risk of fire today"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: true
- short_name: "fire_risk_share_30yrs"
df_field_name: "SHARE_OF_PROPERTIES_AT_RISK_FROM_FIRE_IN_30_YEARS"
long_name: "Share of properties at risk of fire in 30 years"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: true
- long_name: "DOT Travel Disadvantage Index"
short_name: "DOT"
module_name: "travel_composite"
input_geoid_tract_field_name: "GEOID10_TRACT"
load_fields:
- short_name: "travel_burden"
df_field_name: "TRAVEL_BURDEN_FIELD_NAME"
long_name: "DOT Travel Barriers Score"
field_type: float
include_in_tiles: true
include_in_downloadable_files: true
create_percentile: true
- long_name: "National Land Cover Database (NLCD) Lack of Green Space / Nature-Deprived Communities dataset, as compiled by TPL"
short_name: "nlcd_nature_deprived"
module_name: "nlcd_nature_deprived"
input_geoid_tract_field_name: "GEOID10_TRACT"
load_fields:
- short_name: "ncld_eligible"
df_field_name: "ELIGIBLE_FOR_NATURE_DEPRIVED_FIELD_NAME"
long_name: "Does the tract have at least 35 acres in it?"
field_type: bool
include_in_tiles: true
include_in_downloadable_files: true
create_percentile: false
- short_name: "percent_impervious"
df_field_name: "TRACT_PERCENT_IMPERVIOUS_FIELD_NAME"
long_name: "Share of the tract's land area that is covered by impervious surface as a percent"
field_type: percentage
include_in_tiles: true
include_in_downloadable_files: true
create_percentile: true
- short_name: "percent_nonnatural"
df_field_name: "TRACT_PERCENT_NON_NATURAL_FIELD_NAME"
long_name: "Share of the tract's land area that is covered by impervious surface or cropland as a percent"
field_type: percentage
include_in_tiles: true
include_in_downloadable_files: true
create_percentile: true
- short_name: "percent_cropland"
df_field_name: "TRACT_PERCENT_CROPLAND_FIELD_NAME"
long_name: "Share of the tract's land area that is covered by cropland as a percent"
field_type: percentage
include_in_tiles: true
include_in_downloadable_files: true
create_percentile: true
- long_name: "Overlap between Census tract boundaries and Tribal area boundaries."
short_name: "tribal_overlap"
module_name: "tribal_overlap"
input_geoid_tract_field_name: "GEOID10_TRACT"
load_fields:
- short_name: "tribal_count"
df_field_name: "COUNT_OF_TRIBAL_AREAS_IN_TRACT"
long_name: "Number of Tribal areas within Census tract"
field_type: int64
include_in_tiles: true
include_in_downloadable_files: true
create_percentile: false
- short_name: "tribal_percent"
df_field_name: "PERCENT_OF_TRIBAL_AREA_IN_TRACT"
long_name: "Percent of the Census tract that is within Tribal areas"
field_type: float
include_in_tiles: true
include_in_downloadable_files: true
create_percentile: false
number_of_decimals_in_output: 6
- short_name: "tribal_names"
df_field_name: "NAMES_OF_TRIBAL_AREAS_IN_TRACT"
long_name: "Names of Tribal areas within Census tract"
field_type: string
include_in_tiles: true
include_in_downloadable_files: true
- long_name: "CDC Life Expeectancy"
short_name: "cdc_life_expectancy"
module_name: "cdc_life_expectancy"
input_geoid_tract_field_name: "Tract ID"
load_fields:
- short_name: "LLEF"
df_field_name: "LIFE_EXPECTANCY_FIELD_NAME"
long_name: "Life expectancy (years)"
field_type: float
include_in_tiles: false
include_in_downloadable_files: true
create_percentile: false
create_reverse_percentile: true

236
data/data-pipeline/data_pipeline/etl/score/constants.py
View file

@ -2,9 +2,11 @@ import os
from pathlib import Path
from data_pipeline.config import settings
from data_pipeline.score import field_names
## note: to keep map porting "right" fields, keeping descriptors the same.
# Base Paths
DATA_PATH = Path(settings.APP_ROOT) / "data"
TMP_PATH = DATA_PATH / "tmp"
@ -115,7 +117,7 @@ ISLAND_AREAS_EXPLANATION = (
CENSUS_COUNTIES_COLUMNS = ["USPS", "GEOID", "NAME"]
# Drop FIPS codes from map
DROP_FIPS_CODES = ["66", "78"]
DROP_FIPS_CODES = []
# Drop FIPS codes from incrementing
DROP_FIPS_FROM_NON_WTD_THRESHOLDS = "72"
@ -138,7 +140,7 @@ TILES_ROUND_NUM_DECIMALS = 2
# Controlling Tile user experience columns
THRESHOLD_COUNT_TO_SHOW_FIELD_NAME = "THRHLD"
TILES_ISLAND_AREAS_THRESHOLD_COUNT = 3
TILES_PUERTO_RICO_THRESHOLD_COUNT = 4
TILES_PUERTO_RICO_THRESHOLD_COUNT = 10
TILES_NATION_THRESHOLD_COUNT = 21
# Note that the FIPS code is a string
@ -146,6 +148,58 @@ TILES_NATION_THRESHOLD_COUNT = 21
# 60: American Samoa, 66: Guam, 69: N. Mariana Islands, 78: US Virgin Islands
TILES_ISLAND_AREA_FIPS_CODES = ["60", "66", "69", "78"]
TILES_PUERTO_RICO_FIPS_CODE = ["72"]
TILES_ALASKA_AND_HAWAII_FIPS_CODE = ["02", "15"]
TILES_CONTINENTAL_US_FIPS_CODE = [
"01",
"04",
"05",
"06",
"08",
"09",
"10",
"11",
"12",
"13",
"16",
"17",
"18",
"19",
"20",
"21",
"22",
"23",
"24",
"25",
"26",
"27",
"28",
"29",
"30",
"31",
"32",
"33",
"34",
"35",
"36",
"37",
"38",
"39",
"40",
"41",
"42",
"44",
"45",
"46",
"47",
"48",
"49",
"50",
"51",
"53",
"54",
"55",
"56",
]
# Constant to reflect UI Experience version
# "Nation" referring to 50 states and DC is from Census
@ -189,16 +243,17 @@ TILES_SCORE_COLUMNS = {
+ field_names.PERCENTILE_FIELD_SUFFIX: "LIF_PFS",
field_names.LOW_MEDIAN_INCOME_AS_PERCENT_OF_AMI_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "LMI_PFS",
field_names.MEDIAN_HOUSE_VALUE_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "MHVF_PFS",
field_names.PM25_FIELD + field_names.PERCENTILE_FIELD_SUFFIX: "PM25F_PFS",
field_names.HIGH_SCHOOL_ED_FIELD: "HSEF",
field_names.POVERTY_LESS_THAN_100_FPL_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "P100_PFS",
field_names.POVERTY_LESS_THAN_200_FPL_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "P200_PFS",
field_names.POVERTY_LESS_THAN_200_FPL_IMPUTED_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "P200_I_PFS",
field_names.FPL_200_SERIES_IMPUTED_AND_ADJUSTED_DONUTS: "AJDLI_ET",
field_names.LEAD_PAINT_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "LPF_PFS",
field_names.NO_KITCHEN_OR_INDOOR_PLUMBING_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "KP_PFS",
field_names.NPL_FIELD + field_names.PERCENTILE_FIELD_SUFFIX: "NPL_PFS",
field_names.RMP_FIELD + field_names.PERCENTILE_FIELD_SUFFIX: "RMP_PFS",
field_names.TSDF_FIELD + field_names.PERCENTILE_FIELD_SUFFIX: "TSDF_PFS",
@ -208,37 +263,24 @@ TILES_SCORE_COLUMNS = {
+ field_names.PERCENTILE_FIELD_SUFFIX: "UF_PFS",
field_names.WASTEWATER_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "WF_PFS",
field_names.M_WATER: "M_WTR",
field_names.M_WORKFORCE: "M_WKFC",
field_names.M_CLIMATE: "M_CLT",
field_names.M_ENERGY: "M_ENY",
field_names.M_TRANSPORTATION: "M_TRN",
field_names.M_HOUSING: "M_HSG",
field_names.M_POLLUTION: "M_PLN",
field_names.M_HEALTH: "M_HLTH",
field_names.SCORE_M_COMMUNITIES: "SM_C",
field_names.SCORE_M + field_names.PERCENTILE_FIELD_SUFFIX: "SM_PFS",
field_names.EXPECTED_POPULATION_LOSS_RATE_LOW_INCOME_LOW_HIGHER_ED_FIELD: "EPLRLI",
field_names.EXPECTED_AGRICULTURE_LOSS_RATE_LOW_INCOME_LOW_HIGHER_ED_FIELD: "EALRLI",
field_names.EXPECTED_BUILDING_LOSS_RATE_LOW_INCOME_LOW_HIGHER_ED_FIELD: "EBLRLI",
field_names.PM25_EXPOSURE_LOW_INCOME_LOW_HIGHER_ED_FIELD: "PM25LI",
field_names.ENERGY_BURDEN_LOW_INCOME_LOW_HIGHER_ED_FIELD: "EBLI",
field_names.DIESEL_PARTICULATE_MATTER_LOW_INCOME_LOW_HIGHER_ED_FIELD: "DPMLI",
field_names.TRAFFIC_PROXIMITY_LOW_INCOME_LOW_HIGHER_ED_FIELD: "TPLI",
field_names.LEAD_PAINT_MEDIAN_HOUSE_VALUE_LOW_INCOME_LOW_HIGHER_ED_FIELD: "LPMHVLI",
field_names.HOUSING_BURDEN_LOW_INCOME_LOW_HIGHER_ED_FIELD: "HBLI",
field_names.RMP_LOW_INCOME_LOW_HIGHER_ED_FIELD: "RMPLI",
field_names.SUPERFUND_LOW_INCOME_LOW_HIGHER_ED_FIELD: "SFLI",
field_names.HAZARDOUS_WASTE_LOW_INCOME_LOW_HIGHER_ED_FIELD: "HWLI",
field_names.WASTEWATER_DISCHARGE_LOW_INCOME_LOW_HIGHER_ED_FIELD: "WDLI",
field_names.DIABETES_LOW_INCOME_LOW_HIGHER_ED_FIELD: "DLI",
field_names.ASTHMA_LOW_INCOME_LOW_HIGHER_ED_FIELD: "ALI",
field_names.HEART_DISEASE_LOW_INCOME_LOW_HIGHER_ED_FIELD: "HDLI",
field_names.LOW_LIFE_EXPECTANCY_LOW_INCOME_LOW_HIGHER_ED_FIELD: "LLELI",
field_names.LINGUISTIC_ISOLATION_LOW_HS_LOW_HIGHER_ED_FIELD: "LILHSE",
field_names.POVERTY_LOW_HS_LOW_HIGHER_ED_FIELD: "PLHSE",
field_names.LOW_MEDIAN_INCOME_LOW_HS_LOW_HIGHER_ED_FIELD: "LMILHSE",
field_names.UNEMPLOYMENT_LOW_HS_LOW_HIGHER_ED_FIELD: "ULHSE",
field_names.UST_FIELD + field_names.PERCENTILE_FIELD_SUFFIX: "UST_PFS",
field_names.N_WATER: "N_WTR",
field_names.N_WORKFORCE: "N_WKFC",
field_names.N_CLIMATE: "N_CLT",
field_names.N_ENERGY: "N_ENY",
field_names.N_TRANSPORTATION: "N_TRN",
field_names.N_HOUSING: "N_HSG",
field_names.N_POLLUTION: "N_PLN",
field_names.N_HEALTH: "N_HLTH",
# temporarily update this so that it's the Narwhal score that gets visualized on the map
# The NEW final score value INCLUDES the adjacency index.
field_names.FINAL_SCORE_N_BOOLEAN: "SN_C",
field_names.IS_TRIBAL_DAC: "SN_T",
field_names.DIABETES_LOW_INCOME_FIELD: "DLI",
field_names.ASTHMA_LOW_INCOME_FIELD: "ALI",
field_names.POVERTY_LOW_HS_EDUCATION_FIELD: "PLHSE",
field_names.LOW_MEDIAN_INCOME_LOW_HS_EDUCATION_FIELD: "LMILHSE",
field_names.UNEMPLOYMENT_LOW_HS_EDUCATION_FIELD: "ULHSE",
# new booleans only for the environmental factors
field_names.EXPECTED_POPULATION_LOSS_EXCEEDS_PCTILE_THRESHOLD: "EPL_ET",
field_names.EXPECTED_AGRICULTURAL_LOSS_EXCEEDS_PCTILE_THRESHOLD: "EAL_ET",
@ -248,11 +290,14 @@ TILES_SCORE_COLUMNS = {
field_names.DIESEL_EXCEEDS_PCTILE_THRESHOLD: "DS_ET",
field_names.TRAFFIC_PROXIMITY_PCTILE_THRESHOLD: "TP_ET",
field_names.LEAD_PAINT_PROXY_PCTILE_THRESHOLD: "LPP_ET",
field_names.HISTORIC_REDLINING_SCORE_EXCEEDED: "HRS_ET",
field_names.NO_KITCHEN_OR_INDOOR_PLUMBING_PCTILE_THRESHOLD: "KP_ET",
field_names.HOUSING_BURDEN_PCTILE_THRESHOLD: "HB_ET",
field_names.RMP_PCTILE_THRESHOLD: "RMP_ET",
field_names.NPL_PCTILE_THRESHOLD: "NPL_ET",
field_names.TSDF_PCTILE_THRESHOLD: "TSDF_ET",
field_names.WASTEWATER_PCTILE_THRESHOLD: "WD_ET",
field_names.UST_PCTILE_THRESHOLD: "UST_ET",
field_names.DIABETES_PCTILE_THRESHOLD: "DB_ET",
field_names.ASTHMA_PCTILE_THRESHOLD: "A_ET",
field_names.HEART_DISEASE_PCTILE_THRESHOLD: "HD_ET",
@ -278,79 +323,56 @@ TILES_SCORE_COLUMNS = {
field_names.CENSUS_DECENNIAL_UNEMPLOYMENT_FIELD_2009
+ field_names.ISLAND_AREAS_PERCENTILE_ADJUSTMENT_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "IAULHSE_PFS",
field_names.LOW_HS_EDUCATION_LOW_HIGHER_ED_FIELD: "LHE",
field_names.LOW_HS_EDUCATION_FIELD: "LHE",
field_names.ISLAND_AREAS_LOW_HS_EDUCATION_FIELD: "IALHE",
# Percentage of HS Degree completion for Islands
field_names.CENSUS_DECENNIAL_HIGH_SCHOOL_ED_FIELD_2009: "IAHSEF",
field_names.COLLEGE_ATTENDANCE_FIELD: "CA",
field_names.COLLEGE_NON_ATTENDANCE_FIELD: "NCA",
# This is logically equivalent to "non-college greater than 80%"
field_names.COLLEGE_ATTENDANCE_LESS_THAN_20_FIELD: "CA_LT20",
# Booleans for the front end about the types of thresholds exceeded
field_names.CLIMATE_THRESHOLD_EXCEEDED: "M_CLT_EOMI",
field_names.ENERGY_THRESHOLD_EXCEEDED: "M_ENY_EOMI",
field_names.TRAFFIC_THRESHOLD_EXCEEDED: "M_TRN_EOMI",
field_names.HOUSING_THREHSOLD_EXCEEDED: "M_HSG_EOMI",
field_names.POLLUTION_THRESHOLD_EXCEEDED: "M_PLN_EOMI",
field_names.WATER_THRESHOLD_EXCEEDED: "M_WTR_EOMI",
field_names.HEALTH_THRESHOLD_EXCEEDED: "M_HLTH_EOMI",
field_names.WORKFORCE_THRESHOLD_EXCEEDED: "M_WKFC_EOMI",
field_names.CLIMATE_THRESHOLD_EXCEEDED: "N_CLT_EOMI",
field_names.ENERGY_THRESHOLD_EXCEEDED: "N_ENY_EOMI",
field_names.TRAFFIC_THRESHOLD_EXCEEDED: "N_TRN_EOMI",
field_names.HOUSING_THREHSOLD_EXCEEDED: "N_HSG_EOMI",
field_names.POLLUTION_THRESHOLD_EXCEEDED: "N_PLN_EOMI",
field_names.WATER_THRESHOLD_EXCEEDED: "N_WTR_EOMI",
field_names.HEALTH_THRESHOLD_EXCEEDED: "N_HLTH_EOMI",
field_names.WORKFORCE_THRESHOLD_EXCEEDED: "N_WKFC_EOMI",
# These are the booleans for socioeconomic indicators
## this measures low income boolean
field_names.FPL_200_SERIES: "FPL200S",
## Low high school and low higher ed for t&wd
field_names.WORKFORCE_SOCIO_INDICATORS_EXCEEDED: "M_WKFC_EBSI",
## FPL 200 and low higher ed for all others
field_names.FPL_200_AND_COLLEGE_ATTENDANCE_SERIES: "M_EBSI",
field_names.FPL_200_SERIES_IMPUTED_AND_ADJUSTED: "FPL200S",
## Low high school for t&wd
field_names.WORKFORCE_SOCIO_INDICATORS_EXCEEDED: "N_WKFC_EBSI",
field_names.DOT_BURDEN_PCTILE_THRESHOLD: "TD_ET",
field_names.DOT_TRAVEL_BURDEN_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "TD_PFS",
field_names.FUTURE_FLOOD_RISK_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "FLD_PFS",
field_names.FUTURE_WILDFIRE_RISK_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX: "WFR_PFS",
field_names.HIGH_FUTURE_FLOOD_RISK_FIELD: "FLD_ET",
field_names.HIGH_FUTURE_WILDFIRE_RISK_FIELD: "WFR_ET",
field_names.ADJACENT_TRACT_SCORE_ABOVE_DONUT_THRESHOLD: "ADJ_ET",
field_names.TRACT_PERCENT_NON_NATURAL_FIELD_NAME
+ field_names.PERCENTILE_FIELD_SUFFIX: "IS_PFS",
field_names.NON_NATURAL_LOW_INCOME_FIELD_NAME: "IS_ET",
field_names.AML_BOOLEAN_FILLED_IN: "AML_ET",
field_names.ELIGIBLE_FUDS_BINARY_FIELD_NAME: "FUDS_RAW",
field_names.ELIGIBLE_FUDS_FILLED_IN_FIELD_NAME: "FUDS_ET",
field_names.IMPUTED_INCOME_FLAG_FIELD_NAME: "IMP_FLG",
## FPL 200 and low higher ed for all others should no longer be M_EBSI, but rather
## FPL_200 (there is no higher ed in narwhal)
field_names.PERCENT_BLACK_FIELD_NAME: "DM_B",
field_names.PERCENT_AMERICAN_INDIAN_FIELD_NAME: "DM_AI",
field_names.PERCENT_ASIAN_FIELD_NAME: "DM_A",
field_names.PERCENT_HAWAIIAN_FIELD_NAME: "DM_HI",
field_names.PERCENT_TWO_OR_MORE_RACES_FIELD_NAME: "DM_T",
field_names.PERCENT_NON_HISPANIC_WHITE_FIELD_NAME: "DM_W",
field_names.PERCENT_HISPANIC_FIELD_NAME: "DM_H",
field_names.PERCENT_OTHER_RACE_FIELD_NAME: "DM_O",
field_names.PERCENT_AGE_UNDER_10: "AGE_10",
field_names.PERCENT_AGE_10_TO_64: "AGE_MIDDLE",
field_names.PERCENT_AGE_OVER_64: "AGE_OLD",
field_names.COUNT_OF_TRIBAL_AREAS_IN_TRACT_AK: "TA_COUNT_AK",
field_names.COUNT_OF_TRIBAL_AREAS_IN_TRACT_CONUS: "TA_COUNT_C",
field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT: "TA_PERC",
field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT_DISPLAY: "TA_PERC_FE",
}
# columns to round floats to 2 decimals
# TODO refactor to use much smaller subset of fields we DON'T want to round
TILES_SCORE_FLOAT_COLUMNS = [
field_names.DIABETES_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.ASTHMA_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.HEART_DISEASE_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.DIESEL_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.ENERGY_BURDEN_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.EXPECTED_AGRICULTURE_LOSS_RATE_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX,
field_names.EXPECTED_BUILDING_LOSS_RATE_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX,
field_names.EXPECTED_POPULATION_LOSS_RATE_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX,
field_names.HOUSING_BURDEN_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.LOW_LIFE_EXPECTANCY_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.LINGUISTIC_ISO_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.LOW_MEDIAN_INCOME_AS_PERCENT_OF_AMI_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX,
field_names.MEDIAN_HOUSE_VALUE_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.PM25_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.POVERTY_LESS_THAN_100_FPL_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX,
field_names.POVERTY_LESS_THAN_200_FPL_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX,
field_names.LEAD_PAINT_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.NPL_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.RMP_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.TSDF_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.TRAFFIC_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.UNEMPLOYMENT_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
# Percentiles for Island areas' workforce columns
# To be clear: the island areas pull from 2009 census. PR does not.
field_names.LOW_CENSUS_DECENNIAL_AREA_MEDIAN_INCOME_PERCENT_FIELD_2009
+ field_names.PERCENTILE_FIELD_SUFFIX,
field_names.CENSUS_DECENNIAL_POVERTY_LESS_THAN_100_FPL_FIELD_2009
+ field_names.ISLAND_AREAS_PERCENTILE_ADJUSTMENT_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX,
field_names.CENSUS_DECENNIAL_UNEMPLOYMENT_FIELD_2009
+ field_names.ISLAND_AREAS_PERCENTILE_ADJUSTMENT_FIELD
+ field_names.PERCENTILE_FIELD_SUFFIX,
# Island areas HS degree attainment rate
field_names.CENSUS_DECENNIAL_HIGH_SCHOOL_ED_FIELD_2009,
field_names.LOW_HS_EDUCATION_LOW_HIGHER_ED_FIELD,
field_names.ISLAND_AREAS_LOW_HS_EDUCATION_FIELD,
field_names.WASTEWATER_FIELD + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.SCORE_M + field_names.PERCENTILE_FIELD_SUFFIX,
field_names.COLLEGE_NON_ATTENDANCE_FIELD,
field_names.COLLEGE_ATTENDANCE_FIELD,
]

426
data/data-pipeline/data_pipeline/etl/score/etl_score.py
View file

@ -1,17 +1,26 @@
import functools
from collections import namedtuple
from dataclasses import dataclass
from typing import List
import numpy as np
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.score import constants
from data_pipeline.etl.sources.census_acs.etl import CensusACSETL
from data_pipeline.etl.sources.dot_travel_composite.etl import (
TravelCompositeETL,
)
from data_pipeline.etl.sources.eamlis.etl import AbandonedMineETL
from data_pipeline.etl.sources.fsf_flood_risk.etl import FloodRiskETL
from data_pipeline.etl.sources.fsf_wildfire_risk.etl import WildfireRiskETL
from data_pipeline.etl.sources.national_risk_index.etl import (
NationalRiskIndexETL,
)
from data_pipeline.score.score_runner import ScoreRunner
from data_pipeline.etl.sources.nlcd_nature_deprived.etl import NatureDeprivedETL
from data_pipeline.etl.sources.tribal_overlap.etl import TribalOverlapETL
from data_pipeline.etl.sources.us_army_fuds.etl import USArmyFUDS
from data_pipeline.score import field_names
from data_pipeline.etl.score import constants
from data_pipeline.score.score_runner import ScoreRunner
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)
@ -24,7 +33,7 @@ class ScoreETL(ExtractTransformLoad):
# dataframes
self.df: pd.DataFrame
self.ejscreen_df: pd.DataFrame
self.census_df: pd.DataFrame
self.census_acs_df: pd.DataFrame
self.hud_housing_df: pd.DataFrame
self.cdc_places_df: pd.DataFrame
self.census_acs_median_incomes_df: pd.DataFrame
@ -32,18 +41,25 @@ class ScoreETL(ExtractTransformLoad):
self.doe_energy_burden_df: pd.DataFrame
self.national_risk_index_df: pd.DataFrame
self.geocorr_urban_rural_df: pd.DataFrame
self.persistent_poverty_df: pd.DataFrame
self.census_decennial_df: pd.DataFrame
self.census_2010_df: pd.DataFrame
self.child_opportunity_index_df: pd.DataFrame
self.national_tract_df: pd.DataFrame
self.hrs_df: pd.DataFrame
self.dot_travel_disadvantage_df: pd.DataFrame
self.fsf_flood_df: pd.DataFrame
self.fsf_fire_df: pd.DataFrame
self.nature_deprived_df: pd.DataFrame
self.eamlis_df: pd.DataFrame
self.fuds_df: pd.DataFrame
self.tribal_overlap_df: pd.DataFrame
self.ISLAND_DEMOGRAPHIC_BACKFILL_FIELDS: List[str] = []
def extract(self) -> None:
logger.info("Loading data sets from disk.")
# EJSCreen csv Load
ejscreen_csv = (
constants.DATA_PATH / "dataset" / "ejscreen_2019" / "usa.csv"
)
ejscreen_csv = constants.DATA_PATH / "dataset" / "ejscreen" / "usa.csv"
self.ejscreen_df = pd.read_csv(
ejscreen_csv,
dtype={self.GEOID_TRACT_FIELD_NAME: "string"},
@ -51,14 +67,7 @@ class ScoreETL(ExtractTransformLoad):
)
# Load census data
census_csv = (
constants.DATA_PATH / "dataset" / "census_acs_2019" / "usa.csv"
)
self.census_df = pd.read_csv(
census_csv,
dtype={self.GEOID_TRACT_FIELD_NAME: "string"},
low_memory=False,
)
self.census_acs_df = CensusACSETL.get_data_frame()
# Load HUD housing data
hud_housing_csv = (
@ -116,6 +125,27 @@ class ScoreETL(ExtractTransformLoad):
# Load FEMA national risk index data
self.national_risk_index_df = NationalRiskIndexETL.get_data_frame()
# Load DOT Travel Disadvantage
self.dot_travel_disadvantage_df = TravelCompositeETL.get_data_frame()
# Load fire risk data
self.fsf_fire_df = WildfireRiskETL.get_data_frame()
# Load flood risk data
self.fsf_flood_df = FloodRiskETL.get_data_frame()
# Load NLCD Nature-Deprived Communities data
self.nature_deprived_df = NatureDeprivedETL.get_data_frame()
# Load eAMLIS dataset
self.eamlis_df = AbandonedMineETL.get_data_frame()
# Load FUDS dataset
self.fuds_df = USArmyFUDS.get_data_frame()
# Load Tribal overlap dataset
self.tribal_overlap_df = TribalOverlapETL.get_data_frame()
# Load GeoCorr Urban Rural Map
geocorr_urban_rural_csv = (
constants.DATA_PATH / "dataset" / "geocorr" / "usa.csv"
@ -126,16 +156,6 @@ class ScoreETL(ExtractTransformLoad):
low_memory=False,
)
# Load persistent poverty
persistent_poverty_csv = (
constants.DATA_PATH / "dataset" / "persistent_poverty" / "usa.csv"
)
self.persistent_poverty_df = pd.read_csv(
persistent_poverty_csv,
dtype={self.GEOID_TRACT_FIELD_NAME: "string"},
low_memory=False,
)
# Load decennial census data
census_decennial_csv = (
constants.DATA_PATH
@ -159,19 +179,26 @@ class ScoreETL(ExtractTransformLoad):
low_memory=False,
)
# Load COI data
child_opportunity_index_csv = (
constants.DATA_PATH
/ "dataset"
/ "child_opportunity_index"
/ "usa.csv"
# Load HRS data
hrs_csv = (
constants.DATA_PATH / "dataset" / "historic_redlining" / "usa.csv"
)
self.child_opportunity_index_df = pd.read_csv(
child_opportunity_index_csv,
self.hrs_df = pd.read_csv(
hrs_csv,
dtype={self.GEOID_TRACT_FIELD_NAME: "string"},
low_memory=False,
)
national_tract_csv = constants.DATA_CENSUS_CSV_FILE_PATH
self.national_tract_df = pd.read_csv(
national_tract_csv,
names=[self.GEOID_TRACT_FIELD_NAME],
dtype={self.GEOID_TRACT_FIELD_NAME: "string"},
low_memory=False,
header=None,
)
def _join_tract_dfs(self, census_tract_dfs: list) -> pd.DataFrame:
logger.info("Joining Census Tract dataframes")
@ -253,6 +280,7 @@ class ScoreETL(ExtractTransformLoad):
df: pd.DataFrame,
input_column_name: str,
output_column_name_root: str,
drop_tracts: list = None,
ascending: bool = True,
) -> pd.DataFrame:
"""Creates percentiles.
@ -262,98 +290,46 @@ class ScoreETL(ExtractTransformLoad):
E.g., "PM2.5 exposure (percentile)".
This will be for the entire country.
For an "apples-to-apples" comparison of urban tracts to other urban tracts,
and compare rural tracts to other rural tracts.
This percentile will be created and returned as
f"{output_column_name_root}{field_names.PERCENTILE_URBAN_RURAL_FIELD_SUFFIX}".
E.g., "PM2.5 exposure (percentile urban/rural)".
This field exists for every tract, but for urban tracts this value will be the
percentile compared to other urban tracts, and for rural tracts this value
will be the percentile compared to other rural tracts.
Specific methdology:
1. Decide a methodology for confirming whether a tract counts as urban or
rural. Currently in the codebase, we use Geocorr to identify the % rural of
a tract, and mark the tract as rural if the percentage is >50% and urban
otherwise. This may or may not be the right methodology.
2. Once tracts are marked as urban or rural, create one percentile rank
that only ranks urban tracts, and one percentile rank that only ranks rural
tracts.
3. Combine into a single field.
`output_column_name_root` is different from `input_column_name` to enable the
reverse percentile use case. In that use case, `input_column_name` may be
something like "3rd grade reading proficiency" and `output_column_name_root`
may be something like "Low 3rd grade reading proficiency".
"""
if (
output_column_name_root
!= field_names.EXPECTED_AGRICULTURE_LOSS_RATE_FIELD
):
# We have two potential options for assessing how to calculate percentiles.
# For the vast majority of columns, we will simply calculate percentiles overall.
# However, for Linguistic Isolation and Agricultural Value Loss, there exist conditions
# for which we drop out tracts from consideration in the percentile. More details on those
# are below, for them, we provide a list of tracts to not include.
# Because of the fancy transformations below, I have removed the urban / rural percentiles,
# which are now deprecated.
if not drop_tracts:
# Create the "basic" percentile.
## note: I believe this is less performant than if we made a bunch of these PFS columns
## and then concatenated the list. For the refactor!
df[
f"{output_column_name_root}"
f"{field_names.PERCENTILE_FIELD_SUFFIX}"
] = df[input_column_name].rank(pct=True, ascending=ascending)
else:
# For agricultural loss, we are using whether there is value at all to determine percentile and then
# filling places where the value is False with 0
tmp_series = df[input_column_name].where(
~df[field_names.GEOID_TRACT_FIELD].isin(drop_tracts),
np.nan,
)
logger.info(
f"Creating special case column for percentiles from {input_column_name}"
)
df[
f"{output_column_name_root}"
f"{field_names.PERCENTILE_FIELD_SUFFIX}"
] = (
df.where(
df[field_names.AGRICULTURAL_VALUE_BOOL_FIELD].astype(float)
== 1.0
)[input_column_name]
.rank(ascending=ascending, pct=True)
.fillna(
df[field_names.AGRICULTURAL_VALUE_BOOL_FIELD].astype(float)
)
)
] = tmp_series.rank(ascending=ascending, pct=True)
# Create the urban/rural percentiles.
urban_rural_percentile_fields_to_combine = []
for (urban_or_rural_string, urban_heuristic_bool) in [
("urban", True),
("rural", False),
]:
# Create a field with only those values
this_category_only_value_field = (
f"{input_column_name} (value {urban_or_rural_string} only)"
)
df[this_category_only_value_field] = np.where(
df[field_names.URBAN_HEURISTIC_FIELD] == urban_heuristic_bool,
df[input_column_name],
None,
)
# Calculate the percentile for only this category
this_category_only_percentile_field = (
f"{output_column_name_root} "
f"(percentile {urban_or_rural_string} only)"
)
df[this_category_only_percentile_field] = df[
this_category_only_value_field
].rank(
pct=True,
# Set ascending to the parameter value.
ascending=ascending,
)
# Add the field name to this list. Later, we'll combine this list.
urban_rural_percentile_fields_to_combine.append(
this_category_only_percentile_field
)
# Combine both urban and rural into one field:
df[
f"{output_column_name_root}{field_names.PERCENTILE_URBAN_RURAL_FIELD_SUFFIX}"
] = df[urban_rural_percentile_fields_to_combine].mean(
axis=1, skipna=True
)
# Check that "drop tracts" were dropped (quicker than creating a fixture?)
assert df[df[field_names.GEOID_TRACT_FIELD].isin(drop_tracts)][
f"{output_column_name_root}"
f"{field_names.PERCENTILE_FIELD_SUFFIX}"
].isna().sum() == len(drop_tracts), "Not all tracts were dropped"
return df
@ -363,19 +339,25 @@ class ScoreETL(ExtractTransformLoad):
# Join all the data sources that use census tracts
census_tract_dfs = [
self.census_df,
self.census_acs_df,
self.hud_housing_df,
self.cdc_places_df,
self.cdc_life_expectancy_df,
self.doe_energy_burden_df,
self.ejscreen_df,
self.geocorr_urban_rural_df,
self.persistent_poverty_df,
self.national_risk_index_df,
self.census_acs_median_incomes_df,
self.census_decennial_df,
self.census_2010_df,
self.child_opportunity_index_df,
self.hrs_df,
self.dot_travel_disadvantage_df,
self.fsf_flood_df,
self.fsf_fire_df,
self.nature_deprived_df,
self.eamlis_df,
self.fuds_df,
self.tribal_overlap_df,
]
# Sanity check each data frame before merging.
@ -384,8 +366,22 @@ class ScoreETL(ExtractTransformLoad):
census_tract_df = self._join_tract_dfs(census_tract_dfs)
# If GEOID10s are read as numbers instead of strings, the initial 0 is dropped,
# and then we get too many CBG rows (one for 012345 and one for 12345).
# Drop tracts that don't exist in the 2010 tracts
pre_join_len = census_tract_df[field_names.GEOID_TRACT_FIELD].nunique()
census_tract_df = census_tract_df.merge(
self.national_tract_df,
on="GEOID10_TRACT",
how="inner",
)
assert (
census_tract_df.shape[0] <= pre_join_len
), "Join against national tract list ADDED rows"
logger.info(
"Dropped %s tracts not in the 2010 tract data",
pre_join_len
- census_tract_df[field_names.GEOID_TRACT_FIELD].nunique(),
)
# Now sanity-check the merged df.
self._census_tract_df_sanity_check(
@ -405,8 +401,29 @@ class ScoreETL(ExtractTransformLoad):
df[field_names.MEDIAN_INCOME_FIELD] / df[field_names.AMI_FIELD]
)
self.ISLAND_DEMOGRAPHIC_BACKFILL_FIELDS = [
field_names.PERCENT_BLACK_FIELD_NAME
+ field_names.ISLAND_AREA_BACKFILL_SUFFIX,
field_names.PERCENT_AMERICAN_INDIAN_FIELD_NAME
+ field_names.ISLAND_AREA_BACKFILL_SUFFIX,
field_names.PERCENT_ASIAN_FIELD_NAME
+ field_names.ISLAND_AREA_BACKFILL_SUFFIX,
field_names.PERCENT_HAWAIIAN_FIELD_NAME
+ field_names.ISLAND_AREA_BACKFILL_SUFFIX,
field_names.PERCENT_TWO_OR_MORE_RACES_FIELD_NAME
+ field_names.ISLAND_AREA_BACKFILL_SUFFIX,
field_names.PERCENT_NON_HISPANIC_WHITE_FIELD_NAME
+ field_names.ISLAND_AREA_BACKFILL_SUFFIX,
field_names.PERCENT_HISPANIC_FIELD_NAME
+ field_names.ISLAND_AREA_BACKFILL_SUFFIX,
field_names.PERCENT_OTHER_RACE_FIELD_NAME
+ field_names.ISLAND_AREA_BACKFILL_SUFFIX,
]
# Donut columns get added later
numeric_columns = [
field_names.HOUSING_BURDEN_FIELD,
field_names.NO_KITCHEN_OR_INDOOR_PLUMBING_FIELD,
field_names.TOTAL_POP_FIELD,
field_names.MEDIAN_INCOME_AS_PERCENT_OF_STATE_FIELD,
field_names.ASTHMA_FIELD,
@ -453,27 +470,55 @@ class ScoreETL(ExtractTransformLoad):
field_names.CENSUS_UNEMPLOYMENT_FIELD_2010,
field_names.CENSUS_POVERTY_LESS_THAN_100_FPL_FIELD_2010,
field_names.CENSUS_DECENNIAL_TOTAL_POPULATION_FIELD_2009,
field_names.EXTREME_HEAT_FIELD,
field_names.HEALTHY_FOOD_FIELD,
field_names.IMPENETRABLE_SURFACES_FIELD,
# We have to pass this boolean here in order to include it in ag value loss percentiles.
field_names.AGRICULTURAL_VALUE_BOOL_FIELD,
]
field_names.UST_FIELD,
field_names.DOT_TRAVEL_BURDEN_FIELD,
field_names.FUTURE_FLOOD_RISK_FIELD,
field_names.FUTURE_WILDFIRE_RISK_FIELD,
field_names.TRACT_PERCENT_NON_NATURAL_FIELD_NAME,
field_names.POVERTY_LESS_THAN_200_FPL_IMPUTED_FIELD,
field_names.PERCENT_BLACK_FIELD_NAME,
field_names.PERCENT_AMERICAN_INDIAN_FIELD_NAME,
field_names.PERCENT_ASIAN_FIELD_NAME,
field_names.PERCENT_HAWAIIAN_FIELD_NAME,
field_names.PERCENT_TWO_OR_MORE_RACES_FIELD_NAME,
field_names.PERCENT_NON_HISPANIC_WHITE_FIELD_NAME,
field_names.PERCENT_HISPANIC_FIELD_NAME,
field_names.PERCENT_OTHER_RACE_FIELD_NAME,
field_names.PERCENT_AGE_UNDER_10,
field_names.PERCENT_AGE_10_TO_64,
field_names.PERCENT_AGE_OVER_64,
field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT,
field_names.COUNT_OF_TRIBAL_AREAS_IN_TRACT_AK,
field_names.COUNT_OF_TRIBAL_AREAS_IN_TRACT_CONUS,
field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT_DISPLAY,
] + self.ISLAND_DEMOGRAPHIC_BACKFILL_FIELDS
non_numeric_columns = [
self.GEOID_TRACT_FIELD_NAME,
field_names.PERSISTENT_POVERTY_FIELD,
field_names.TRACT_ELIGIBLE_FOR_NONNATURAL_THRESHOLD,
field_names.AGRICULTURAL_VALUE_BOOL_FIELD,
field_names.NAMES_OF_TRIBAL_AREAS_IN_TRACT,
]
boolean_columns = [
field_names.AML_BOOLEAN,
field_names.IMPUTED_INCOME_FLAG_FIELD_NAME,
field_names.ELIGIBLE_FUDS_BINARY_FIELD_NAME,
field_names.HISTORIC_REDLINING_SCORE_EXCEEDED,
field_names.IS_TRIBAL_DAC,
]
# For some columns, high values are "good", so we want to reverse the percentile
# so that high values are "bad" and any scoring logic can still check if it's
# >= some threshold.
# Note that we must use dataclass here instead of namedtuples on account of pylint
# TODO: Add more fields here.
# https://github.com/usds/justice40-tool/issues/970
ReversePercentile = namedtuple(
typename="ReversePercentile",
field_names=["field_name", "low_field_name"],
)
@dataclass
class ReversePercentile:
field_name: str
low_field_name: str
reverse_percentiles = [
# This dictionary follows the format:
# <field name> : <field name for low values>
@ -481,10 +526,6 @@ class ScoreETL(ExtractTransformLoad):
# This low field will not exist yet, it is only calculated for the
# percentile.
# TODO: This will come from the YAML dataset config
ReversePercentile(
field_name=field_names.READING_FIELD,
low_field_name=field_names.LOW_READING_FIELD,
),
ReversePercentile(
field_name=field_names.MEDIAN_INCOME_AS_PERCENT_OF_AMI_FIELD,
low_field_name=field_names.LOW_MEDIAN_INCOME_AS_PERCENT_OF_AMI_FIELD,
@ -503,40 +544,90 @@ class ScoreETL(ExtractTransformLoad):
non_numeric_columns
+ numeric_columns
+ [rp.field_name for rp in reverse_percentiles]
+ boolean_columns
)
df_copy = df[columns_to_keep].copy()
assert len(numeric_columns) == len(
set(numeric_columns)
), "You have a double-entered column in the numeric columns list"
df_copy[numeric_columns] = df_copy[numeric_columns].apply(pd.to_numeric)
# coerce all booleans to bools preserving nan character
# since this is a boolean, need to use `None`
for col in boolean_columns:
tmp = df_copy[col].copy()
df_copy[col] = np.where(tmp.notna(), tmp.astype(bool), None)
logger.info(f"{col} contains {df_copy[col].isna().sum()} nulls.")
# Convert all columns to numeric and do math
# Note that we have a few special conditions here and we handle them explicitly.
# For *Linguistic Isolation*, we do NOT want to include Puerto Rico in the percentile
# calculation. This is because linguistic isolation as a category doesn't make much sense
# in Puerto Rico, where Spanish is a recognized language. Thus, we construct a list
# of tracts to drop from the percentile calculation.
#
# For *Expected Agricultural Loss*, we only want to include in the percentile tracts
# in which there is some agricultural value. This helps us adjust the data such that we have
# the ability to discern which tracts truly are at the 90th percentile, since many tracts have 0 value.
#
# For *Non-Natural Space*, we may only want to include tracts that have at least 35 acreas, I think. This will
# get rid of tracts that we think are aberrations statistically. Right now, we have left this out
# pending ground-truthing.
#
# For *Traffic Barriers*, we want to exclude low population tracts, which may have high burden because they are
# low population alone. We set this low population constant in the if statement.
for numeric_column in numeric_columns:
drop_tracts = []
if (
numeric_column
== field_names.EXPECTED_AGRICULTURE_LOSS_RATE_FIELD
):
drop_tracts = df_copy[
~df_copy[field_names.AGRICULTURAL_VALUE_BOOL_FIELD]
.astype(bool)
.fillna(False)
][field_names.GEOID_TRACT_FIELD].to_list()
logger.info(
f"Dropping {len(drop_tracts)} tracts from Agricultural Value Loss"
)
elif numeric_column == field_names.LINGUISTIC_ISO_FIELD:
drop_tracts = df_copy[
# 72 is the FIPS code for Puerto Rico
df_copy[field_names.GEOID_TRACT_FIELD].str.startswith("72")
][field_names.GEOID_TRACT_FIELD].to_list()
logger.info(
f"Dropping {len(drop_tracts)} tracts from Linguistic Isolation"
)
elif numeric_column in [
field_names.DOT_TRAVEL_BURDEN_FIELD,
field_names.EXPECTED_POPULATION_LOSS_RATE_FIELD,
]:
# Not having any people appears to be correlated with transit burden, but also doesn't represent
# on the ground need. For now, we remove these tracts from the percentile calculation.ß
# Similarly, we want to exclude low population tracts from FEMA's index
low_population = 20
drop_tracts = df_copy[
df_copy[field_names.TOTAL_POP_FIELD].fillna(0)
<= low_population
][field_names.GEOID_TRACT_FIELD].to_list()
logger.info(
f"Dropping {len(drop_tracts)} tracts from DOT traffic burden"
)
df_copy = self._add_percentiles_to_df(
df=df_copy,
input_column_name=numeric_column,
# For this use case, the input name and output name root are the same.
output_column_name_root=numeric_column,
ascending=True,
drop_tracts=drop_tracts,
)
# Min-max normalization:
# (
# Observed value
# - minimum of all values
# )
# divided by
# (
# Maximum of all values
# - minimum of all values
# )
min_value = df_copy[numeric_column].min(skipna=True)
max_value = df_copy[numeric_column].max(skipna=True)
df_copy[f"{numeric_column}{field_names.MIN_MAX_FIELD_SUFFIX}"] = (
df_copy[numeric_column] - min_value
) / (max_value - min_value)
# Create reversed percentiles for these fields
for reverse_percentile in reverse_percentiles:
# Calculate reverse percentiles
@ -566,6 +657,32 @@ class ScoreETL(ExtractTransformLoad):
return df_copy
@staticmethod
def _get_island_areas(df: pd.DataFrame) -> pd.Series:
return (
df[field_names.GEOID_TRACT_FIELD]
.str[:2]
.isin(constants.TILES_ISLAND_AREA_FIPS_CODES)
)
def _backfill_island_demographics(self, df: pd.DataFrame) -> pd.DataFrame:
logger.info("Backfilling island demographic data")
island_index = self._get_island_areas(df)
for backfill_field_name in self.ISLAND_DEMOGRAPHIC_BACKFILL_FIELDS:
actual_field_name = backfill_field_name.replace(
field_names.ISLAND_AREA_BACKFILL_SUFFIX, ""
)
df.loc[island_index, actual_field_name] = df.loc[
island_index, backfill_field_name
]
df = df.drop(columns=self.ISLAND_DEMOGRAPHIC_BACKFILL_FIELDS)
df.loc[island_index, field_names.TOTAL_POP_FIELD] = df.loc[
island_index, field_names.COMBINED_CENSUS_TOTAL_POPULATION_2010
]
return df
def transform(self) -> None:
logger.info("Transforming Score Data")
@ -575,8 +692,13 @@ class ScoreETL(ExtractTransformLoad):
# calculate scores
self.df = ScoreRunner(df=self.df).calculate_scores()
# We add island demographic data since it doesn't matter to the score anyway
self.df = self._backfill_island_demographics(self.df)
def load(self) -> None:
logger.info("Saving Score CSV")
logger.info(
f"Saving Score CSV to {constants.DATA_SCORE_CSV_FULL_FILE_PATH}."
)
constants.DATA_SCORE_CSV_FULL_DIR.mkdir(parents=True, exist_ok=True)
self.df.to_csv(constants.DATA_SCORE_CSV_FULL_FILE_PATH, index=False)

56
data/data-pipeline/data_pipeline/etl/score/etl_score_geo.py
View file

@ -1,24 +1,20 @@
import concurrent.futures
import math
import os
import geopandas as gpd
import numpy as np
import pandas as pd
import geopandas as gpd
from data_pipeline.content.schemas.download_schemas import CSVConfig
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.score import constants
from data_pipeline.etl.sources.census.etl_utils import (
check_census_data_source,
)
from data_pipeline.etl.score.etl_utils import check_score_data_source
from data_pipeline.etl.sources.census.etl_utils import check_census_data_source
from data_pipeline.score import field_names
from data_pipeline.content.schemas.download_schemas import CSVConfig
from data_pipeline.utils import (
get_module_logger,
zip_files,
load_yaml_dict_from_file,
load_dict_from_yaml_object_fields,
)
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import load_dict_from_yaml_object_fields
from data_pipeline.utils import load_yaml_dict_from_file
from data_pipeline.utils import zip_files
logger = get_module_logger(__name__)
@ -41,23 +37,25 @@ class GeoScoreETL(ExtractTransformLoad):
self.SCORE_CSV_PATH = self.DATA_PATH / "score" / "csv"
self.TILE_SCORE_CSV = self.SCORE_CSV_PATH / "tiles" / "usa.csv"
self.DATA_SOURCE = data_source
self.CENSUS_USA_GEOJSON = (
self.DATA_PATH / "census" / "geojson" / "us.json"
)
# Import the shortened name for Score M percentile ("SM_PFS") that's used on the
# tiles.
# Import the shortened name for Score N to be used on tiles.
# We should no longer be using PFS
## TODO: We really should not have this any longer changing
self.TARGET_SCORE_SHORT_FIELD = constants.TILES_SCORE_COLUMNS[
field_names.SCORE_M + field_names.PERCENTILE_FIELD_SUFFIX
field_names.FINAL_SCORE_N_BOOLEAN
]
self.TARGET_SCORE_RENAME_TO = "M_SCORE"
self.TARGET_SCORE_RENAME_TO = "SCORE"
# Import the shortened name for tract ("GTF") that's used on the tiles.
self.TRACT_SHORT_FIELD = constants.TILES_SCORE_COLUMNS[
field_names.GEOID_TRACT_FIELD
]
self.GEOMETRY_FIELD_NAME = "geometry"
self.LAND_FIELD_NAME = "ALAND10"
# We will adjust this upwards while there is some fractional value
# in the score. This is a starting value.
@ -84,17 +82,28 @@ class GeoScoreETL(ExtractTransformLoad):
)
logger.info("Reading US GeoJSON (~6 minutes)")
self.geojson_usa_df = gpd.read_file(
full_geojson_usa_df = gpd.read_file(
self.CENSUS_USA_GEOJSON,
dtype={self.GEOID_FIELD_NAME: "string"},
usecols=[self.GEOID_FIELD_NAME, self.GEOMETRY_FIELD_NAME],
usecols=[
self.GEOID_FIELD_NAME,
self.GEOMETRY_FIELD_NAME,
self.LAND_FIELD_NAME,
],
low_memory=False,
)
# We only want to keep tracts to visualize that have non-0 land
self.geojson_usa_df = full_geojson_usa_df[
full_geojson_usa_df[self.LAND_FIELD_NAME] > 0
]
logger.info("Reading score CSV")
self.score_usa_df = pd.read_csv(
self.TILE_SCORE_CSV,
dtype={self.TRACT_SHORT_FIELD: "string"},
dtype={
self.TRACT_SHORT_FIELD: str,
},
low_memory=False,
)
@ -134,7 +143,7 @@ class GeoScoreETL(ExtractTransformLoad):
columns={self.TARGET_SCORE_SHORT_FIELD: self.TARGET_SCORE_RENAME_TO}
)
logger.info("Converting to geojson into tracts")
logger.info("Converting geojson into geodf with tracts")
usa_tracts = gpd.GeoDataFrame(
usa_tracts,
columns=[
@ -270,8 +279,10 @@ class GeoScoreETL(ExtractTransformLoad):
# Create separate threads to run each write to disk.
def write_high_to_file():
logger.info("Writing usa-high (~9 minutes)")
self.geojson_score_usa_high.to_file(
filename=self.SCORE_HIGH_GEOJSON, driver="GeoJSON"
filename=self.SCORE_HIGH_GEOJSON,
driver="GeoJSON",
)
logger.info("Completed writing usa-high")
@ -294,7 +305,6 @@ class GeoScoreETL(ExtractTransformLoad):
pd.Series(codebook)
.reset_index()
.rename(
# kept as strings because no downstream impacts
columns={
0: internal_column_name_field,
"index": shapefile_column_field,

124
data/data-pipeline/data_pipeline/etl/score/etl_score_post.py
View file

@ -1,35 +1,29 @@
from pathlib import Path
import json
from numpy import float64
from pathlib import Path
import numpy as np
import pandas as pd
from data_pipeline.content.schemas.download_schemas import (
CSVConfig,
CodebookConfig,
ExcelConfig,
)
from data_pipeline.content.schemas.download_schemas import CodebookConfig
from data_pipeline.content.schemas.download_schemas import CSVConfig
from data_pipeline.content.schemas.download_schemas import ExcelConfig
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.score.etl_utils import floor_series, create_codebook
from data_pipeline.utils import (
get_module_logger,
zip_files,
load_yaml_dict_from_file,
column_list_from_yaml_object_fields,
load_dict_from_yaml_object_fields,
)
from data_pipeline.etl.score.etl_utils import create_codebook
from data_pipeline.etl.score.etl_utils import floor_series
from data_pipeline.etl.sources.census.etl_utils import check_census_data_source
from data_pipeline.score import field_names
from data_pipeline.utils import column_list_from_yaml_object_fields
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import load_dict_from_yaml_object_fields
from data_pipeline.utils import load_yaml_dict_from_file
from data_pipeline.utils import zip_files
from numpy import float64
from data_pipeline.etl.sources.census.etl_utils import (
check_census_data_source,
)
from . import constants
logger = get_module_logger(__name__)
# Define the DAC variable
DISADVANTAGED_COMMUNITIES_FIELD = field_names.SCORE_M_COMMUNITIES
DISADVANTAGED_COMMUNITIES_FIELD = field_names.SCORE_N_COMMUNITIES
class PostScoreETL(ExtractTransformLoad):
@ -45,7 +39,6 @@ class PostScoreETL(ExtractTransformLoad):
self.input_counties_df: pd.DataFrame
self.input_states_df: pd.DataFrame
self.input_score_df: pd.DataFrame
self.input_national_tract_df: pd.DataFrame
self.output_score_county_state_merged_df: pd.DataFrame
self.output_score_tiles_df: pd.DataFrame
@ -92,7 +85,9 @@ class PostScoreETL(ExtractTransformLoad):
def _extract_score(self, score_path: Path) -> pd.DataFrame:
logger.info("Reading Score CSV")
df = pd.read_csv(
score_path, dtype={self.GEOID_TRACT_FIELD_NAME: "string"}
score_path,
dtype={self.GEOID_TRACT_FIELD_NAME: "string"},
low_memory=False,
)
# Convert total population to an int
@ -102,18 +97,6 @@ class PostScoreETL(ExtractTransformLoad):
return df
def _extract_national_tract(
self, national_tract_path: Path
) -> pd.DataFrame:
logger.info("Reading national tract file")
return pd.read_csv(
national_tract_path,
names=[self.GEOID_TRACT_FIELD_NAME],
dtype={self.GEOID_TRACT_FIELD_NAME: "string"},
low_memory=False,
header=None,
)
def extract(self) -> None:
logger.info("Starting Extraction")
@ -136,9 +119,6 @@ class PostScoreETL(ExtractTransformLoad):
self.input_score_df = self._extract_score(
constants.DATA_SCORE_CSV_FULL_FILE_PATH
)
self.input_national_tract_df = self._extract_national_tract(
constants.DATA_CENSUS_CSV_FILE_PATH
)
def _transform_counties(
self, initial_counties_df: pd.DataFrame
@ -185,7 +165,6 @@ class PostScoreETL(ExtractTransformLoad):
def _create_score_data(
self,
national_tract_df: pd.DataFrame,
counties_df: pd.DataFrame,
states_df: pd.DataFrame,
score_df: pd.DataFrame,
@ -217,28 +196,11 @@ class PostScoreETL(ExtractTransformLoad):
right_on=self.STATE_CODE_COLUMN,
how="left",
)
# check if there are census tracts without score
logger.info("Removing tract rows without score")
# merge census tracts with score
merged_df = national_tract_df.merge(
score_county_state_merged,
on=self.GEOID_TRACT_FIELD_NAME,
how="left",
)
# recast population to integer
score_county_state_merged["Total population"] = (
merged_df["Total population"].fillna(0).astype(int)
)
de_duplicated_df = merged_df.dropna(
subset=[DISADVANTAGED_COMMUNITIES_FIELD]
)
assert score_county_merged[
self.GEOID_TRACT_FIELD_NAME
].is_unique, "Merging state/county data introduced duplicate rows"
# set the score to the new df
return de_duplicated_df
return score_county_state_merged
def _create_tile_data(
self,
@ -254,8 +216,8 @@ class PostScoreETL(ExtractTransformLoad):
tiles_score_column_titles
].copy()
# Currently, we do not want USVI or Guam on the map, so this will drop all
# rows with the FIPS codes (first two digits of the census tract)
# We may not want some states/territories on the map, so this will drop all
# rows with those FIPS codes (first two digits of the census tract)
logger.info(
f"Dropping specified FIPS codes from tile data: {constants.DROP_FIPS_CODES}"
)
@ -269,16 +231,15 @@ class PostScoreETL(ExtractTransformLoad):
score_tiles = score_tiles[
~score_tiles[field_names.GEOID_TRACT_FIELD].isin(tracts_to_drop)
]
score_tiles[constants.TILES_SCORE_FLOAT_COLUMNS] = score_tiles[
constants.TILES_SCORE_FLOAT_COLUMNS
].apply(
func=lambda series: floor_series(
series=series,
number_of_decimals=constants.TILES_ROUND_NUM_DECIMALS,
),
axis=0,
)
float_cols = [
col
for col, col_dtype in score_tiles.dtypes.items()
if col_dtype == np.dtype("float64")
]
scale_factor = 10**constants.TILES_ROUND_NUM_DECIMALS
score_tiles[float_cols] = (
score_tiles[float_cols] * scale_factor
).apply(np.floor) / scale_factor
logger.info("Adding fields for island areas and Puerto Rico")
# The below operation constructs variables for the front end.
@ -427,7 +388,6 @@ class PostScoreETL(ExtractTransformLoad):
transformed_score = self._transform_score(self.input_score_df)
output_score_county_state_merged_df = self._create_score_data(
self.input_national_tract_df,
transformed_counties,
transformed_states,
transformed_score,
@ -521,8 +481,6 @@ class PostScoreETL(ExtractTransformLoad):
score_tiles_df.to_csv(tile_score_path, index=False, encoding="utf-8")
def _load_downloadable_zip(self, downloadable_info_path: Path) -> None:
logger.info("Saving Downloadable CSV")
downloadable_info_path.mkdir(parents=True, exist_ok=True)
csv_path = constants.SCORE_DOWNLOADABLE_CSV_FILE_PATH
excel_path = constants.SCORE_DOWNLOADABLE_EXCEL_FILE_PATH
@ -583,6 +541,22 @@ class PostScoreETL(ExtractTransformLoad):
"fields"
],
)
assert codebook_df["csv_label"].equals(codebook_df["excel_label"]), (
"CSV and Excel differ. If that's intentional, "
"remove this assertion. Otherwise, fix it."
)
# Check the codebook to make sure it matches the download files
assert not set(codebook_df["csv_label"].dropna()).difference(
downloadable_df.columns
), "Codebook is missing columns from downloadable files"
assert (
len(
downloadable_df.columns.difference(
set(codebook_df["csv_label"])
)
)
== 0
), "Codebook has columns the downloadable files do not"
# load codebook to disk
codebook_df.to_csv(codebook_path, index=False)

122
data/data-pipeline/data_pipeline/etl/score/etl_utils.py
View file

@ -1,16 +1,21 @@
import os
import sys
from pathlib import Path
import typing
from collections import namedtuple
from pathlib import Path
import numpy as np
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.utils import (
download_file_from_url,
get_module_logger,
)
from data_pipeline.etl.score.constants import TILES_ALASKA_AND_HAWAII_FIPS_CODE
from data_pipeline.etl.score.constants import TILES_CONTINENTAL_US_FIPS_CODE
from data_pipeline.etl.score.constants import TILES_ISLAND_AREA_FIPS_CODES
from data_pipeline.etl.score.constants import TILES_PUERTO_RICO_FIPS_CODE
from data_pipeline.etl.sources.census.etl_utils import get_state_fips_codes
from data_pipeline.score import field_names
from data_pipeline.utils import download_file_from_url
from data_pipeline.utils import get_module_logger
from . import constants
logger = get_module_logger(__name__)
@ -91,7 +96,7 @@ def floor_series(series: pd.Series, number_of_decimals: int) -> pd.Series:
if series.isin(unacceptable_values).any():
series.replace(mapping, regex=False, inplace=True)
multiplication_factor = 10 ** number_of_decimals
multiplication_factor = 10**number_of_decimals
# In order to safely cast NaNs
# First coerce series to float type: series.astype(float)
@ -305,3 +310,106 @@ def create_codebook(
return merged_codebook_df[constants.CODEBOOK_COLUMNS].rename(
columns={constants.CEJST_SCORE_COLUMN_NAME: "Description"}
)
# pylint: disable=too-many-arguments
def compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes: typing.List[str],
continental_us_expected: bool = True,
alaska_and_hawaii_expected: bool = True,
puerto_rico_expected: bool = True,
island_areas_expected: bool = True,
additional_fips_codes_not_expected: typing.List[str] = None,
dataset_name: str = None,
) -> None:
"""Check whether a list of state/territory FIPS codes match expectations.
Args:
actual_state_fips_codes (List of str): Actual state codes observed in data
continental_us_expected (bool, optional): Do you expect the continental nation
(DC & states except for Alaska and Hawaii) to be represented in data?
alaska_and_hawaii_expected (bool, optional): Do you expect Alaska and Hawaii
to be represented in the data? Note: if only *1* of Alaska and Hawaii are
not expected to be included, do not use this argument -- instead,
use `additional_fips_codes_not_expected` for the 1 state you expected to
be missing.
puerto_rico_expected (bool, optional): Do you expect PR to be represented in data?
island_areas_expected (bool, optional): Do you expect Island Areas to be represented in
data?
additional_fips_codes_not_expected (List of str, optional): Additional state codes
not expected in the data. For example, the data may be known to be missing
data from Maine and Wisconsin.
dataset_name (str, optional): The name of the data set, used only in printing an
error message. (This is helpful for debugging during parallel etl runs.)
Returns:
None: Does not return any values.
Raises:
ValueError: if lists do not match expectations.
"""
# Setting default argument of [] here to avoid mutability problems.
if additional_fips_codes_not_expected is None:
additional_fips_codes_not_expected = []
# Cast input to a set.
actual_state_fips_codes_set = set(actual_state_fips_codes)
# Start with the list of all FIPS codes for all states and territories.
expected_states_set = set(get_state_fips_codes(settings.DATA_PATH))
# If continental US is not expected to be included, remove it from the
# expected states set.
if not continental_us_expected:
expected_states_set = expected_states_set.difference(
TILES_CONTINENTAL_US_FIPS_CODE
)
# If both Alaska and Hawaii are not expected to be included, remove them from the
# expected states set.
# Note: if only *1* of Alaska and Hawaii are not expected to be included,
# do not use this argument -- instead, use `additional_fips_codes_not_expected`
# for the 1 state you expected to be missing.
if not alaska_and_hawaii_expected:
expected_states_set = expected_states_set.difference(
TILES_ALASKA_AND_HAWAII_FIPS_CODE
)
# If Puerto Rico is not expected to be included, remove it from the expected
# states set.
if not puerto_rico_expected:
expected_states_set = expected_states_set.difference(
TILES_PUERTO_RICO_FIPS_CODE
)
# If island areas are not expected to be included, remove them from the expected
# states set.
if not island_areas_expected:
expected_states_set = expected_states_set.difference(
TILES_ISLAND_AREA_FIPS_CODES
)
# If additional FIPS codes are not expected to be included, remove them from the
# expected states set.
expected_states_set = expected_states_set.difference(
additional_fips_codes_not_expected
)
dataset_name_phrase = (
f" for dataset `{dataset_name}`" if dataset_name is not None else ""
)
if expected_states_set != actual_state_fips_codes_set:
raise ValueError(
f"The states and territories in the data{dataset_name_phrase} are not "
f"as expected.\n"
"FIPS state codes expected that are not present in the data:\n"
f"{sorted(list(expected_states_set - actual_state_fips_codes_set))}\n"
"FIPS state codes in the data that were not expected:\n"
f"{sorted(list(actual_state_fips_codes_set - expected_states_set))}\n"
)
else:
logger.info(
"Data matches expected state and territory representation"
f"{dataset_name_phrase}."
)

8
data/data-pipeline/data_pipeline/etl/score/schemas/datasets.py
View file

@ -1,6 +1,8 @@
from dataclasses import dataclass, field
from dataclasses import dataclass
from dataclasses import field
from enum import Enum
from typing import List, Optional
from typing import List
from typing import Optional
class FieldType(Enum):
@ -77,7 +79,7 @@ class DatasetsConfig:
long_name: str
short_name: str
module_name: str
input_geoid_tract_field_name: str
load_fields: List[LoadField]
input_geoid_tract_field_name: Optional[str] = None
datasets: List[Dataset]

3
data/data-pipeline/data_pipeline/etl/score/tests/conftest.py
View file

@ -5,7 +5,8 @@ from pathlib import Path
import pandas as pd
import pytest
from data_pipeline import config
from data_pipeline.etl.score import etl_score_post, tests
from data_pipeline.etl.score import etl_score_post
from data_pipeline.etl.score import tests
from data_pipeline.etl.score.etl_score_post import PostScoreETL

6
data/data-pipeline/data_pipeline/etl/score/tests/sample_data/score_data_initial.csv
View file

File diff suppressed because one or more lines are too long

2
data/data-pipeline/data_pipeline/etl/score/tests/sample_data/state_data_initial.csv
View file

@ -1,4 +1,4 @@
fips,state_name,state_abbreviation,region,division
01,Alabama,AL,South,East South Central
02,Alaska,AK,West,Pacific
04,Arizona,AZ,West,Mountain
04,Arizona,AZ,West,Mountain

1 fips state_name state_abbreviation region division
2 01 Alabama AL South East South Central
3 02 Alaska AK West Pacific
4 04 Arizona AZ West Mountain

BIN
data/data-pipeline/data_pipeline/etl/score/tests/snapshots/downloadable_data_expected.pkl
View file

Binary file not shown.

BIN
data/data-pipeline/data_pipeline/etl/score/tests/snapshots/score_data_expected.pkl
View file

Binary file not shown.

BIN
data/data-pipeline/data_pipeline/etl/score/tests/snapshots/score_transformed_expected.pkl
View file

Binary file not shown.

BIN
data/data-pipeline/data_pipeline/etl/score/tests/snapshots/tile_data_expected.pkl
View file

Binary file not shown.

184
data/data-pipeline/data_pipeline/etl/score/tests/test_etl_utils.py
View file

@ -1,7 +1,9 @@
import pandas as pd
import numpy as np
import pandas as pd
import pytest
from data_pipeline.etl.score.etl_utils import (
compare_to_list_of_expected_state_fips_codes,
)
from data_pipeline.etl.score.etl_utils import floor_series
@ -70,3 +72,181 @@ def test_floor_series():
match="Argument series must be of type pandas series, not of type list.",
):
floor_series(invalid_type, number_of_decimals=3)
def test_compare_to_list_of_expected_state_fips_codes():
# Has every state/territory/DC code
fips_codes_test_1 = [
"01",
"02",
"04",
"05",
"06",
"08",
"09",
"10",
"11",
"12",
"13",
"15",
"16",
"17",
"18",
"19",
"20",
"21",
"22",
"23",
"24",
"25",
"26",
"27",
"28",
"29",
"30",
"31",
"32",
"33",
"34",
"35",
"36",
"37",
"38",
"39",
"40",
"41",
"42",
"44",
"45",
"46",
"47",
"48",
"49",
"50",
"51",
"53",
"54",
"55",
"56",
"60",
"66",
"69",
"72",
"78",
]
# Should not raise any errors
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=fips_codes_test_1
)
# Should raise error because Puerto Rico is not expected
with pytest.raises(ValueError) as exception_info:
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=fips_codes_test_1,
puerto_rico_expected=False,
)
partial_expected_error_message = (
"FIPS state codes in the data that were not expected:\n['72']\n"
)
assert partial_expected_error_message in str(exception_info.value)
# Should raise error because Island Areas are not expected
with pytest.raises(ValueError) as exception_info:
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=fips_codes_test_1,
island_areas_expected=False,
)
partial_expected_error_message = (
"FIPS state codes in the data that were not expected:\n"
"['60', '66', '69', '78']\n"
)
assert partial_expected_error_message in str(exception_info.value)
# List missing PR and Guam
fips_codes_test_2 = [x for x in fips_codes_test_1 if x not in ["66", "72"]]
# Should raise error because all Island Areas and PR are expected
with pytest.raises(ValueError) as exception_info:
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=fips_codes_test_2,
)
partial_expected_error_message = (
"FIPS state codes expected that are not present in the data:\n"
"['66', '72']\n"
)
assert partial_expected_error_message in str(exception_info.value)
# Missing Maine and Wisconsin
fips_codes_test_3 = [x for x in fips_codes_test_1 if x not in ["23", "55"]]
# Should raise error because Maine and Wisconsin are expected
with pytest.raises(ValueError) as exception_info:
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=fips_codes_test_3,
)
partial_expected_error_message = (
"FIPS state codes expected that are not present in the data:\n"
"['23', '55']\n"
)
assert partial_expected_error_message in str(exception_info.value)
# Should not raise error because Maine and Wisconsin are expected to be missing
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=fips_codes_test_3,
additional_fips_codes_not_expected=["23", "55"],
)
# Missing the continental & AK/HI nation
fips_codes_test_4 = [
"60",
"66",
"69",
"72",
"78",
]
# Should raise error because the nation is expected
with pytest.raises(ValueError) as exception_info:
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=fips_codes_test_4,
)
partial_expected_error_message = (
"FIPS state codes expected that are not present in the data:\n"
"['01', '02', '04', '05', '06', '08', '09', '10', '11', '12', '13', '15', '16', "
"'17', '18', '19', '20', '21', '22', '23', '24', '25', '26', '27', '28', '29', "
"'30', '31', '32', '33', '34', '35', '36', '37', '38', '39', '40', '41', '42', "
"'44', '45', '46', '47', '48', '49', '50', '51', '53', '54', '55', '56']"
)
assert partial_expected_error_message in str(exception_info.value)
# Should not raise error because continental US and AK/HI is not to be missing
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=fips_codes_test_4,
continental_us_expected=False,
alaska_and_hawaii_expected=False,
)
# Missing Hawaii but not Alaska
fips_codes_test_5 = [x for x in fips_codes_test_1 if x not in ["15"]]
# Should raise error because both Hawaii and Alaska are expected
with pytest.raises(ValueError) as exception_info:
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=fips_codes_test_5,
alaska_and_hawaii_expected=True,
)
partial_expected_error_message = (
"FIPS state codes expected that are not present in the data:\n"
"['15']\n"
)
assert partial_expected_error_message in str(exception_info.value)
# Should work as expected
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=fips_codes_test_5,
alaska_and_hawaii_expected=True,
additional_fips_codes_not_expected=["15"],
)

9
data/data-pipeline/data_pipeline/etl/score/tests/test_score_post.py
View file

@ -1,14 +1,11 @@
# pylint: disable=W0212
## Above disables warning about access to underscore-prefixed methods
from importlib import reload
from pathlib import Path
import pandas.api.types as ptypes
import pandas.testing as pdt
from data_pipeline.content.schemas.download_schemas import (
CSVConfig,
)
from data_pipeline.content.schemas.download_schemas import CSVConfig
from data_pipeline.etl.score import constants
from data_pipeline.utils import load_yaml_dict_from_file
@ -67,14 +64,12 @@ def test_transform_score(etl, score_data_initial, score_transformed_expected):
# pylint: disable=too-many-arguments
def test_create_score_data(
etl,
national_tract_df,
counties_transformed_expected,
states_transformed_expected,
score_transformed_expected,
score_data_expected,
):
score_data_actual = etl._create_score_data(
national_tract_df,
counties_transformed_expected,
states_transformed_expected,
score_transformed_expected,

3
data/data-pipeline/data_pipeline/etl/sources/calenviroscreen/etl.py
View file

@ -1,8 +1,7 @@
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
logger = get_module_logger(__name__)

130
data/data-pipeline/data_pipeline/etl/sources/cdc_life_expectancy/etl.py
View file

@ -1,58 +1,148 @@
import pathlib
from pathlib import Path
import pandas as pd
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger, download_file_from_url
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.etl.score.etl_utils import (
compare_to_list_of_expected_state_fips_codes,
)
from data_pipeline.score import field_names
from data_pipeline.utils import download_file_from_url
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
logger = get_module_logger(__name__)
class CDCLifeExpectancy(ExtractTransformLoad):
GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
PUERTO_RICO_EXPECTED_IN_DATA = False
NAME = "cdc_life_expectancy"
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
USA_FILE_URL = f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/cdc_file_expectancy/US_A.CSV"
else:
USA_FILE_URL: str = "https://ftp.cdc.gov/pub/Health_Statistics/NCHS/Datasets/NVSS/USALEEP/CSV/US_A.CSV"
LOAD_YAML_CONFIG: bool = False
LIFE_EXPECTANCY_FIELD_NAME = "Life expectancy (years)"
INPUT_GEOID_TRACT_FIELD_NAME = "Tract ID"
STATES_MISSING_FROM_USA_FILE = ["23", "55"]
# For some reason, LEEP does not include Maine or Wisconsin in its "All of
# USA" file. Load these separately.
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
WISCONSIN_FILE_URL: str = f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/cdc_file_expectancy/WI_A.CSV"
MAINE_FILE_URL: str = f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/cdc_file_expectancy/ME_A.CSV"
else:
WISCONSIN_FILE_URL: str = "https://ftp.cdc.gov/pub/Health_Statistics/NCHS/Datasets/NVSS/USALEEP/CSV/WI_A.CSV"
MAINE_FILE_URL: str = "https://ftp.cdc.gov/pub/Health_Statistics/NCHS/Datasets/NVSS/USALEEP/CSV/ME_A.CSV"
TRACT_INPUT_COLUMN_NAME = "Tract ID"
STATE_INPUT_COLUMN_NAME = "STATE2KX"
raw_df: pd.DataFrame
output_df: pd.DataFrame
def __init__(self):
self.FILE_URL: str = "https://ftp.cdc.gov/pub/Health_Statistics/NCHS/Datasets/NVSS/USALEEP/CSV/US_A.CSV"
self.OUTPUT_PATH: Path = (
self.DATA_PATH / "dataset" / "cdc_life_expectancy"
)
self.TRACT_INPUT_COLUMN_NAME = "Tract ID"
self.LIFE_EXPECTANCY_FIELD_NAME = "Life expectancy (years)"
# Constants for output
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
self.LIFE_EXPECTANCY_FIELD_NAME,
field_names.LIFE_EXPECTANCY_FIELD,
]
self.raw_df: pd.DataFrame
self.output_df: pd.DataFrame
def extract(self) -> None:
logger.info("Starting data download.")
download_file_name = (
self.get_tmp_path() / "cdc_life_expectancy" / "usa.csv"
)
def _download_and_prep_data(
self, file_url: str, download_file_name: pathlib.Path
) -> pd.DataFrame:
download_file_from_url(
file_url=self.FILE_URL,
file_url=file_url,
download_file_name=download_file_name,
verify=True,
)
self.raw_df = pd.read_csv(
df = pd.read_csv(
filepath_or_buffer=download_file_name,
dtype={
# The following need to remain as strings for all of their digits, not get converted to numbers.
self.TRACT_INPUT_COLUMN_NAME: "string",
self.STATE_INPUT_COLUMN_NAME: "string",
},
low_memory=False,
)
return df
def extract(self) -> None:
logger.info("Starting data download.")
all_usa_raw_df = self._download_and_prep_data(
file_url=self.USA_FILE_URL,
download_file_name=self.get_tmp_path() / "US_A.CSV",
)
# Check which states are missing
states_in_life_expectancy_usa_file = list(
all_usa_raw_df[self.STATE_INPUT_COLUMN_NAME].unique()
)
# Expect that PR, Island Areas, and Maine/Wisconsin are missing
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=states_in_life_expectancy_usa_file,
continental_us_expected=self.CONTINENTAL_US_EXPECTED_IN_DATA,
puerto_rico_expected=self.PUERTO_RICO_EXPECTED_IN_DATA,
island_areas_expected=self.ISLAND_AREAS_EXPECTED_IN_DATA,
additional_fips_codes_not_expected=self.STATES_MISSING_FROM_USA_FILE,
)
logger.info("Downloading data for Maine")
maine_raw_df = self._download_and_prep_data(
file_url=self.MAINE_FILE_URL,
download_file_name=self.get_tmp_path() / "maine.csv",
)
logger.info("Downloading data for Wisconsin")
wisconsin_raw_df = self._download_and_prep_data(
file_url=self.WISCONSIN_FILE_URL,
download_file_name=self.get_tmp_path() / "wisconsin.csv",
)
combined_df = pd.concat(
objs=[all_usa_raw_df, maine_raw_df, wisconsin_raw_df],
ignore_index=True,
verify_integrity=True,
axis=0,
)
states_in_combined_df = list(
combined_df[self.STATE_INPUT_COLUMN_NAME].unique()
)
# Expect that PR and Island Areas are the only things now missing
compare_to_list_of_expected_state_fips_codes(
actual_state_fips_codes=states_in_combined_df,
continental_us_expected=self.CONTINENTAL_US_EXPECTED_IN_DATA,
puerto_rico_expected=self.PUERTO_RICO_EXPECTED_IN_DATA,
island_areas_expected=self.ISLAND_AREAS_EXPECTED_IN_DATA,
additional_fips_codes_not_expected=[],
)
# Save the updated version
self.raw_df = combined_df
def transform(self) -> None:
logger.info("Starting DOE energy burden transform.")
logger.info("Starting CDC life expectancy transform.")
self.output_df = self.raw_df.rename(
columns={
"e(0)": self.LIFE_EXPECTANCY_FIELD_NAME,
"e(0)": field_names.LIFE_EXPECTANCY_FIELD,
self.TRACT_INPUT_COLUMN_NAME: self.GEOID_TRACT_FIELD_NAME,
}
)

52
data/data-pipeline/data_pipeline/etl/sources/cdc_places/etl.py
View file

@ -1,20 +1,45 @@
import pandas as pd
import typing
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger, download_file_from_url
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.score import field_names
from data_pipeline.utils import download_file_from_url
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
logger = get_module_logger(__name__)
class CDCPlacesETL(ExtractTransformLoad):
NAME = "cdc_places"
GEO_LEVEL: ValidGeoLevel = ValidGeoLevel.CENSUS_TRACT
PUERTO_RICO_EXPECTED_IN_DATA = False
CDC_GEOID_FIELD_NAME = "LocationID"
CDC_VALUE_FIELD_NAME = "Data_Value"
CDC_MEASURE_FIELD_NAME = "Measure"
def __init__(self):
self.OUTPUT_PATH = self.DATA_PATH / "dataset" / "cdc_places"
self.CDC_PLACES_URL = "https://chronicdata.cdc.gov/api/views/cwsq-ngmh/rows.csv?accessType=DOWNLOAD"
self.CDC_GEOID_FIELD_NAME = "LocationID"
self.CDC_VALUE_FIELD_NAME = "Data_Value"
self.CDC_MEASURE_FIELD_NAME = "Measure"
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
self.CDC_PLACES_URL = (
f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/"
"cdc_places/PLACES__Local_Data_for_Better_Health__Census_Tract_Data_2021_release.csv"
)
else:
self.CDC_PLACES_URL = "https://chronicdata.cdc.gov/api/views/cwsq-ngmh/rows.csv?accessType=DOWNLOAD"
self.COLUMNS_TO_KEEP: typing.List[str] = [
self.GEOID_TRACT_FIELD_NAME,
field_names.DIABETES_FIELD,
field_names.ASTHMA_FIELD,
field_names.HEART_DISEASE_FIELD,
field_names.CANCER_FIELD,
field_names.HEALTH_INSURANCE_FIELD,
field_names.PHYS_HEALTH_NOT_GOOD_FIELD,
]
self.df: pd.DataFrame
@ -22,9 +47,7 @@ class CDCPlacesETL(ExtractTransformLoad):
logger.info("Starting to download 520MB CDC Places file.")
file_path = download_file_from_url(
file_url=self.CDC_PLACES_URL,
download_file_name=self.get_tmp_path()
/ "cdc_places"
/ "census_tract.csv",
download_file_name=self.get_tmp_path() / "census_tract.csv",
)
self.df = pd.read_csv(
@ -42,7 +65,6 @@ class CDCPlacesETL(ExtractTransformLoad):
inplace=True,
errors="raise",
)
# Note: Puerto Rico not included.
self.df = self.df.pivot(
index=self.GEOID_TRACT_FIELD_NAME,
@ -65,12 +87,4 @@ class CDCPlacesETL(ExtractTransformLoad):
)
# Make the index (the census tract ID) a column, not the index.
self.df.reset_index(inplace=True)
def load(self) -> None:
logger.info("Saving CDC Places Data")
# mkdir census
self.OUTPUT_PATH.mkdir(parents=True, exist_ok=True)
self.df.to_csv(path_or_buf=self.OUTPUT_PATH / "usa.csv", index=False)
self.output_df = self.df.reset_index()

4
data/data-pipeline/data_pipeline/etl/sources/cdc_svi_index/README.md
View file

@ -53,7 +53,7 @@ For SVI 2018, the authors also included two adjunct variables, 1) 2014-2018 ACS
**Important Notes**
1. Tracts with zero estimates for the total population (N = 645 for the U.S.) were removed during the ranking process. These tracts were added back to the SVI databases after ranking.
1. Tracts with zero estimates for the total population (N = 645 for the U.S.) were removed during the ranking process. These tracts were added back to the SVI databases after ranking.
2. The TOTPOP field value is 0, but the percentile ranking fields (RPL_THEME1, RPL_THEME2, RPL_THEME3, RPL_THEME4, and RPL_THEMES) were set to -999.
@ -66,4 +66,4 @@ here: https://www.census.gov/programs-surveys/acs/data/variance-tables.html.
For selected ACS 5-year Detailed Tables, “Users can calculate margins of error for aggregated data by using the variance replicates. Unlike available approximation formulas, this method results in an exact margin of error by using the covariance term.”
MOEs are _not_ included nor considered during this data processing nor for the scoring comparison tool.
MOEs are _not_ included nor considered during this data processing nor for the scoring comparison tool.

14
data/data-pipeline/data_pipeline/etl/sources/cdc_svi_index/etl.py
View file

@ -1,9 +1,9 @@
import pandas as pd
import numpy as np
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
logger = get_module_logger(__name__)
@ -17,7 +17,13 @@ class CDCSVIIndex(ExtractTransformLoad):
def __init__(self):
self.OUTPUT_PATH = self.DATA_PATH / "dataset" / "cdc_svi_index"
self.CDC_SVI_INDEX_URL = "https://svi.cdc.gov/Documents/Data/2018_SVI_Data/CSV/SVI2018_US.csv"
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
self.CDC_SVI_INDEX_URL = (
f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/"
"cdc_svi_index/SVI2018_US.csv"
)
else:
self.CDC_SVI_INDEX_URL = "https://svi.cdc.gov/Documents/Data/2018_SVI_Data/CSV/SVI2018_US.csv"
self.CDC_RPL_THEMES_THRESHOLD = 0.90

21
data/data-pipeline/data_pipeline/etl/sources/census/etl.py
View file

@ -3,12 +3,12 @@ import json
import subprocess
from enum import Enum
from pathlib import Path
import geopandas as gpd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger, unzip_file_from_url
from data_pipeline.etl.sources.census.etl_utils import get_state_fips_codes
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import unzip_file_from_url
logger = get_module_logger(__name__)
@ -20,19 +20,20 @@ class GeoFileType(Enum):
class CensusETL(ExtractTransformLoad):
SHP_BASE_PATH = ExtractTransformLoad.DATA_PATH / "census" / "shp"
GEOJSON_BASE_PATH = ExtractTransformLoad.DATA_PATH / "census" / "geojson"
CSV_BASE_PATH = ExtractTransformLoad.DATA_PATH / "census" / "csv"
GEOJSON_PATH = ExtractTransformLoad.DATA_PATH / "census" / "geojson"
NATIONAL_TRACT_CSV_PATH = CSV_BASE_PATH / "us.csv"
NATIONAL_TRACT_JSON_PATH = GEOJSON_BASE_PATH / "us.json"
GEOID_TRACT_FIELD_NAME: str = "GEOID10_TRACT"
def __init__(self):
self.SHP_BASE_PATH = self.DATA_PATH / "census" / "shp"
self.GEOJSON_BASE_PATH = self.DATA_PATH / "census" / "geojson"
self.CSV_BASE_PATH = self.DATA_PATH / "census" / "csv"
# the fips_states_2010.csv is generated from data here
# https://www.census.gov/geographies/reference-files/time-series/geo/tallies.html
self.STATE_FIPS_CODES = get_state_fips_codes(self.DATA_PATH)
self.GEOJSON_PATH = self.DATA_PATH / "census" / "geojson"
self.TRACT_PER_STATE: dict = {} # in-memory dict per state
self.TRACT_NATIONAL: list = [] # in-memory global list
self.NATIONAL_TRACT_CSV_PATH = self.CSV_BASE_PATH / "us.csv"
self.NATIONAL_TRACT_JSON_PATH = self.GEOJSON_BASE_PATH / "us.json"
self.GEOID_TRACT_FIELD_NAME: str = "GEOID10_TRACT"
def _path_for_fips_file(
self, fips_code: str, file_type: GeoFileType

12
data/data-pipeline/data_pipeline/etl/sources/census/etl_utils.py
View file

@ -5,13 +5,11 @@ from pathlib import Path
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.utils import (
get_module_logger,
remove_all_dirs_from_dir,
remove_files_from_dir,
unzip_file_from_url,
zip_directory,
)
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import remove_all_dirs_from_dir
from data_pipeline.utils import remove_files_from_dir
from data_pipeline.utils import unzip_file_from_url
from data_pipeline.utils import zip_directory
logger = get_module_logger(__name__)

417
data/data-pipeline/data_pipeline/etl/sources/census_acs/etl.py
View file

@ -1,22 +1,33 @@
import pandas as pd
import os
from collections import namedtuple
import geopandas as gpd
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.sources.census_acs.etl_imputations import (
calculate_income_measures,
)
from data_pipeline.etl.sources.census_acs.etl_utils import (
retrieve_census_acs_data,
)
from data_pipeline.utils import get_module_logger
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import unzip_file_from_url
logger = get_module_logger(__name__)
# because now there is a requirement for the us.json, this will port from
# AWS when a local copy does not exist.
CENSUS_DATA_S3_URL = settings.AWS_JUSTICE40_DATASOURCES_URL + "/census.zip"
class CensusACSETL(ExtractTransformLoad):
def __init__(self):
self.ACS_YEAR = 2019
self.OUTPUT_PATH = (
self.DATA_PATH / "dataset" / f"census_acs_{self.ACS_YEAR}"
)
NAME = "census_acs"
ACS_YEAR = 2019
MINIMUM_POPULATION_REQUIRED_FOR_IMPUTATION = 1
def __init__(self):
self.TOTAL_UNEMPLOYED_FIELD = "B23025_005E"
self.TOTAL_IN_LABOR_FORCE = "B23025_003E"
self.EMPLOYMENT_FIELDS = [
@ -59,6 +70,23 @@ class CensusACSETL(ExtractTransformLoad):
self.POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME = (
"Percent of individuals < 200% Federal Poverty Line"
)
self.IMPUTED_POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME = (
"Percent of individuals < 200% Federal Poverty Line, imputed"
)
self.ADJUSTED_POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME = (
"Adjusted percent of individuals < 200% Federal Poverty Line"
)
self.ADJUSTED_AND_IMPUTED_POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME_PRELIMINARY = (
"Preliminary adjusted percent of individuals < 200% Federal Poverty Line,"
+ " imputed"
)
self.ADJUSTED_AND_IMPUTED_POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME = (
"Adjusted percent of individuals < 200% Federal Poverty Line,"
+ " imputed"
)
self.MEDIAN_HOUSE_VALUE_FIELD = "B25077_001E"
self.MEDIAN_HOUSE_VALUE_FIELD_NAME = (
@ -136,6 +164,10 @@ class CensusACSETL(ExtractTransformLoad):
"Percent enrollment in college or graduate school"
)
self.IMPUTED_COLLEGE_ATTENDANCE_FIELD = (
"Percent enrollment in college or graduate school, imputed"
)
self.COLLEGE_NON_ATTENDANCE_FIELD = "Percent of population not currently enrolled in college or graduate school"
self.RE_FIELDS = [
@ -153,19 +185,25 @@ class CensusACSETL(ExtractTransformLoad):
"B03002_003E",
"B03003_001E",
"B03003_003E",
"B02001_007E", # "Some other race alone"
]
# Name output demographics fields.
self.BLACK_FIELD_NAME = "Black or African American alone"
self.AMERICAN_INDIAN_FIELD_NAME = (
"American Indian and Alaska Native alone"
)
self.ASIAN_FIELD_NAME = "Asian alone"
self.HAWAIIAN_FIELD_NAME = "Native Hawaiian and Other Pacific alone"
self.TWO_OR_MORE_RACES_FIELD_NAME = "Two or more races"
self.NON_HISPANIC_WHITE_FIELD_NAME = "Non-Hispanic White"
self.BLACK_FIELD_NAME = "Black or African American"
self.AMERICAN_INDIAN_FIELD_NAME = "American Indian / Alaska Native"
self.ASIAN_FIELD_NAME = "Asian"
self.HAWAIIAN_FIELD_NAME = "Native Hawaiian or Pacific"
self.TWO_OR_MORE_RACES_FIELD_NAME = "two or more races"
self.NON_HISPANIC_WHITE_FIELD_NAME = "White"
self.HISPANIC_FIELD_NAME = "Hispanic or Latino"
# Note that `other` is lowercase because the whole field will show up in the download
# file as "Percent other races"
self.OTHER_RACE_FIELD_NAME = "other races"
self.TOTAL_RACE_POPULATION_FIELD_NAME = (
"Total population surveyed on racial data"
)
# Name output demographics fields.
self.RE_OUTPUT_FIELDS = [
self.BLACK_FIELD_NAME,
self.AMERICAN_INDIAN_FIELD_NAME,
@ -174,32 +212,133 @@ class CensusACSETL(ExtractTransformLoad):
self.TWO_OR_MORE_RACES_FIELD_NAME,
self.NON_HISPANIC_WHITE_FIELD_NAME,
self.HISPANIC_FIELD_NAME,
self.OTHER_RACE_FIELD_NAME,
]
self.PERCENT_PREFIX = "Percent "
# Note: this field does double-duty here. It's used as the total population
# within the age questions.
# It's also what EJScreen used as their variable for total population in the
# census tract, so we use it similarly.
# See p. 83 of https://www.epa.gov/sites/default/files/2021-04/documents/ejscreen_technical_document.pdf
self.TOTAL_POPULATION_FROM_AGE_TABLE = "B01001_001E" # Estimate!!Total:
self.AGE_INPUT_FIELDS = [
self.TOTAL_POPULATION_FROM_AGE_TABLE,
"B01001_003E", # Estimate!!Total:!!Male:!!Under 5 years
"B01001_004E", # Estimate!!Total:!!Male:!!5 to 9 years
"B01001_005E", # Estimate!!Total:!!Male:!!10 to 14 years
"B01001_006E", # Estimate!!Total:!!Male:!!15 to 17 years
"B01001_007E", # Estimate!!Total:!!Male:!!18 and 19 years
"B01001_008E", # Estimate!!Total:!!Male:!!20 years
"B01001_009E", # Estimate!!Total:!!Male:!!21 years
"B01001_010E", # Estimate!!Total:!!Male:!!22 to 24 years
"B01001_011E", # Estimate!!Total:!!Male:!!25 to 29 years
"B01001_012E", # Estimate!!Total:!!Male:!!30 to 34 years
"B01001_013E", # Estimate!!Total:!!Male:!!35 to 39 years
"B01001_014E", # Estimate!!Total:!!Male:!!40 to 44 years
"B01001_015E", # Estimate!!Total:!!Male:!!45 to 49 years
"B01001_016E", # Estimate!!Total:!!Male:!!50 to 54 years
"B01001_017E", # Estimate!!Total:!!Male:!!55 to 59 years
"B01001_018E", # Estimate!!Total:!!Male:!!60 and 61 years
"B01001_019E", # Estimate!!Total:!!Male:!!62 to 64 years
"B01001_020E", # Estimate!!Total:!!Male:!!65 and 66 years
"B01001_021E", # Estimate!!Total:!!Male:!!67 to 69 years
"B01001_022E", # Estimate!!Total:!!Male:!!70 to 74 years
"B01001_023E", # Estimate!!Total:!!Male:!!75 to 79 years
"B01001_024E", # Estimate!!Total:!!Male:!!80 to 84 years
"B01001_025E", # Estimate!!Total:!!Male:!!85 years and over
"B01001_027E", # Estimate!!Total:!!Female:!!Under 5 years
"B01001_028E", # Estimate!!Total:!!Female:!!5 to 9 years
"B01001_029E", # Estimate!!Total:!!Female:!!10 to 14 years
"B01001_030E", # Estimate!!Total:!!Female:!!15 to 17 years
"B01001_031E", # Estimate!!Total:!!Female:!!18 and 19 years
"B01001_032E", # Estimate!!Total:!!Female:!!20 years
"B01001_033E", # Estimate!!Total:!!Female:!!21 years
"B01001_034E", # Estimate!!Total:!!Female:!!22 to 24 years
"B01001_035E", # Estimate!!Total:!!Female:!!25 to 29 years
"B01001_036E", # Estimate!!Total:!!Female:!!30 to 34 years
"B01001_037E", # Estimate!!Total:!!Female:!!35 to 39 years
"B01001_038E", # Estimate!!Total:!!Female:!!40 to 44 years
"B01001_039E", # Estimate!!Total:!!Female:!!45 to 49 years
"B01001_040E", # Estimate!!Total:!!Female:!!50 to 54 years
"B01001_041E", # Estimate!!Total:!!Female:!!55 to 59 years
"B01001_042E", # Estimate!!Total:!!Female:!!60 and 61 years
"B01001_043E", # Estimate!!Total:!!Female:!!62 to 64 years
"B01001_044E", # Estimate!!Total:!!Female:!!65 and 66 years
"B01001_045E", # Estimate!!Total:!!Female:!!67 to 69 years
"B01001_046E", # Estimate!!Total:!!Female:!!70 to 74 years
"B01001_047E", # Estimate!!Total:!!Female:!!75 to 79 years
"B01001_048E", # Estimate!!Total:!!Female:!!80 to 84 years
"B01001_049E", # Estimate!!Total:!!Female:!!85 years and over
]
self.AGE_OUTPUT_FIELDS = [
field_names.PERCENT_AGE_UNDER_10,
field_names.PERCENT_AGE_10_TO_64,
field_names.PERCENT_AGE_OVER_64,
]
self.STATE_GEOID_FIELD_NAME = "GEOID2"
self.COLUMNS_TO_KEEP = (
[
self.GEOID_TRACT_FIELD_NAME,
field_names.TOTAL_POP_FIELD,
self.UNEMPLOYED_FIELD_NAME,
self.LINGUISTIC_ISOLATION_FIELD_NAME,
self.MEDIAN_INCOME_FIELD_NAME,
self.POVERTY_LESS_THAN_100_PERCENT_FPL_FIELD_NAME,
self.POVERTY_LESS_THAN_150_PERCENT_FPL_FIELD_NAME,
self.POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME,
self.IMPUTED_POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME,
self.MEDIAN_HOUSE_VALUE_FIELD_NAME,
self.HIGH_SCHOOL_ED_FIELD,
self.COLLEGE_ATTENDANCE_FIELD,
self.COLLEGE_NON_ATTENDANCE_FIELD,
self.IMPUTED_COLLEGE_ATTENDANCE_FIELD,
field_names.IMPUTED_INCOME_FLAG_FIELD_NAME,
]
+ self.RE_OUTPUT_FIELDS
+ [self.PERCENT_PREFIX + field for field in self.RE_OUTPUT_FIELDS]
+ [
field_names.PERCENT_PREFIX + field
for field in self.RE_OUTPUT_FIELDS
]
+ self.AGE_OUTPUT_FIELDS
+ [
field_names.POVERTY_LESS_THAN_200_FPL_FIELD,
field_names.POVERTY_LESS_THAN_200_FPL_IMPUTED_FIELD,
]
)
self.df: pd.DataFrame
# pylint: disable=too-many-arguments
def _merge_geojson(
self,
df: pd.DataFrame,
usa_geo_df: gpd.GeoDataFrame,
geoid_field: str = "GEOID10",
geometry_field: str = "geometry",
state_code_field: str = "STATEFP10",
county_code_field: str = "COUNTYFP10",
) -> gpd.GeoDataFrame:
usa_geo_df[geoid_field] = (
usa_geo_df[geoid_field].astype(str).str.zfill(11)
)
return gpd.GeoDataFrame(
df.merge(
usa_geo_df[
[
geoid_field,
geometry_field,
state_code_field,
county_code_field,
]
],
left_on=[self.GEOID_TRACT_FIELD_NAME],
right_on=[geoid_field],
)
)
def extract(self) -> None:
# Define the variables to retrieve
variables = (
@ -213,6 +352,7 @@ class CensusACSETL(ExtractTransformLoad):
+ self.EDUCATIONAL_FIELDS
+ self.RE_FIELDS
+ self.COLLEGE_ATTENDANCE_FIELDS
+ self.AGE_INPUT_FIELDS
)
self.df = retrieve_census_acs_data(
@ -227,12 +367,37 @@ class CensusACSETL(ExtractTransformLoad):
df = self.df
# Rename two fields.
# Here we join the geometry of the US to the dataframe so that we can impute
# The income of neighbors. first this looks locally; if there's no local
# geojson file for all of the US, this will read it off of S3
logger.info("Reading in geojson for the country")
if not os.path.exists(
self.DATA_PATH / "census" / "geojson" / "us.json"
):
logger.info("Fetching Census data from AWS S3")
unzip_file_from_url(
CENSUS_DATA_S3_URL,
self.DATA_PATH / "tmp",
self.DATA_PATH,
)
geo_df = gpd.read_file(
self.DATA_PATH / "census" / "geojson" / "us.json",
)
df = self._merge_geojson(
df=df,
usa_geo_df=geo_df,
)
# Rename some fields.
df = df.rename(
columns={
self.MEDIAN_HOUSE_VALUE_FIELD: self.MEDIAN_HOUSE_VALUE_FIELD_NAME,
self.MEDIAN_INCOME_FIELD: self.MEDIAN_INCOME_FIELD_NAME,
}
self.TOTAL_POPULATION_FROM_AGE_TABLE: field_names.TOTAL_POP_FIELD,
},
errors="raise",
)
# Handle null values for various fields, which are `-666666666`.
@ -318,38 +483,101 @@ class CensusACSETL(ExtractTransformLoad):
)
# Calculate some demographic information.
df[self.BLACK_FIELD_NAME] = df["B02001_003E"]
df[self.AMERICAN_INDIAN_FIELD_NAME] = df["B02001_004E"]
df[self.ASIAN_FIELD_NAME] = df["B02001_005E"]
df[self.HAWAIIAN_FIELD_NAME] = df["B02001_006E"]
df[self.TWO_OR_MORE_RACES_FIELD_NAME] = df["B02001_008E"]
df[self.NON_HISPANIC_WHITE_FIELD_NAME] = df["B03002_003E"]
df[self.HISPANIC_FIELD_NAME] = df["B03003_003E"]
# Calculate demographics as percent
df[self.PERCENT_PREFIX + self.BLACK_FIELD_NAME] = (
df["B02001_003E"] / df["B02001_001E"]
)
df[self.PERCENT_PREFIX + self.AMERICAN_INDIAN_FIELD_NAME] = (
df["B02001_004E"] / df["B02001_001E"]
)
df[self.PERCENT_PREFIX + self.ASIAN_FIELD_NAME] = (
df["B02001_005E"] / df["B02001_001E"]
)
df[self.PERCENT_PREFIX + self.HAWAIIAN_FIELD_NAME] = (
df["B02001_006E"] / df["B02001_001E"]
)
df[self.PERCENT_PREFIX + self.TWO_OR_MORE_RACES_FIELD_NAME] = (
df["B02001_008E"] / df["B02001_001E"]
)
df[self.PERCENT_PREFIX + self.NON_HISPANIC_WHITE_FIELD_NAME] = (
df["B03002_003E"] / df["B03002_001E"]
)
df[self.PERCENT_PREFIX + self.HISPANIC_FIELD_NAME] = (
df["B03003_003E"] / df["B03003_001E"]
df = df.rename(
columns={
"B02001_003E": self.BLACK_FIELD_NAME,
"B02001_004E": self.AMERICAN_INDIAN_FIELD_NAME,
"B02001_005E": self.ASIAN_FIELD_NAME,
"B02001_006E": self.HAWAIIAN_FIELD_NAME,
"B02001_008E": self.TWO_OR_MORE_RACES_FIELD_NAME,
"B03002_003E": self.NON_HISPANIC_WHITE_FIELD_NAME,
"B03003_003E": self.HISPANIC_FIELD_NAME,
"B02001_007E": self.OTHER_RACE_FIELD_NAME,
"B02001_001E": self.TOTAL_RACE_POPULATION_FIELD_NAME,
},
errors="raise",
)
# Calculate college attendance:
for race_field_name in self.RE_OUTPUT_FIELDS:
df[field_names.PERCENT_PREFIX + race_field_name] = (
df[race_field_name] / df[self.TOTAL_RACE_POPULATION_FIELD_NAME]
)
# First value is the `age bucket`, and the second value is a list of all fields
# that will be summed in the calculations of the total population in that age
# bucket.
age_bucket_and_its_sum_columns = [
(
field_names.PERCENT_AGE_UNDER_10,
[
"B01001_003E", # Estimate!!Total:!!Male:!!Under 5 years
"B01001_004E", # Estimate!!Total:!!Male:!!5 to 9 years
"B01001_027E", # Estimate!!Total:!!Female:!!Under 5 years
"B01001_028E", # Estimate!!Total:!!Female:!!5 to 9 years
],
),
(
field_names.PERCENT_AGE_10_TO_64,
[
"B01001_005E", # Estimate!!Total:!!Male:!!10 to 14 years
"B01001_006E", # Estimate!!Total:!!Male:!!15 to 17 years
"B01001_007E", # Estimate!!Total:!!Male:!!18 and 19 years
"B01001_008E", # Estimate!!Total:!!Male:!!20 years
"B01001_009E", # Estimate!!Total:!!Male:!!21 years
"B01001_010E", # Estimate!!Total:!!Male:!!22 to 24 years
"B01001_011E", # Estimate!!Total:!!Male:!!25 to 29 years
"B01001_012E", # Estimate!!Total:!!Male:!!30 to 34 years
"B01001_013E", # Estimate!!Total:!!Male:!!35 to 39 years
"B01001_014E", # Estimate!!Total:!!Male:!!40 to 44 years
"B01001_015E", # Estimate!!Total:!!Male:!!45 to 49 years
"B01001_016E", # Estimate!!Total:!!Male:!!50 to 54 years
"B01001_017E", # Estimate!!Total:!!Male:!!55 to 59 years
"B01001_018E", # Estimate!!Total:!!Male:!!60 and 61 years
"B01001_019E", # Estimate!!Total:!!Male:!!62 to 64 years
"B01001_029E", # Estimate!!Total:!!Female:!!10 to 14 years
"B01001_030E", # Estimate!!Total:!!Female:!!15 to 17 years
"B01001_031E", # Estimate!!Total:!!Female:!!18 and 19 years
"B01001_032E", # Estimate!!Total:!!Female:!!20 years
"B01001_033E", # Estimate!!Total:!!Female:!!21 years
"B01001_034E", # Estimate!!Total:!!Female:!!22 to 24 years
"B01001_035E", # Estimate!!Total:!!Female:!!25 to 29 years
"B01001_036E", # Estimate!!Total:!!Female:!!30 to 34 years
"B01001_037E", # Estimate!!Total:!!Female:!!35 to 39 years
"B01001_038E", # Estimate!!Total:!!Female:!!40 to 44 years
"B01001_039E", # Estimate!!Total:!!Female:!!45 to 49 years
"B01001_040E", # Estimate!!Total:!!Female:!!50 to 54 years
"B01001_041E", # Estimate!!Total:!!Female:!!55 to 59 years
"B01001_042E", # Estimate!!Total:!!Female:!!60 and 61 years
"B01001_043E", # Estimate!!Total:!!Female:!!62 to 64 years
],
),
(
field_names.PERCENT_AGE_OVER_64,
[
"B01001_020E", # Estimate!!Total:!!Male:!!65 and 66 years
"B01001_021E", # Estimate!!Total:!!Male:!!67 to 69 years
"B01001_022E", # Estimate!!Total:!!Male:!!70 to 74 years
"B01001_023E", # Estimate!!Total:!!Male:!!75 to 79 years
"B01001_024E", # Estimate!!Total:!!Male:!!80 to 84 years
"B01001_025E", # Estimate!!Total:!!Male:!!85 years and over
"B01001_044E", # Estimate!!Total:!!Female:!!65 and 66 years
"B01001_045E", # Estimate!!Total:!!Female:!!67 to 69 years
"B01001_046E", # Estimate!!Total:!!Female:!!70 to 74 years
"B01001_047E", # Estimate!!Total:!!Female:!!75 to 79 years
"B01001_048E", # Estimate!!Total:!!Female:!!80 to 84 years
"B01001_049E", # Estimate!!Total:!!Female:!!85 years and over
],
),
]
# For each age bucket, sum the relevant columns and calculate the total
# percentage.
for age_bucket, sum_columns in age_bucket_and_its_sum_columns:
df[age_bucket] = (
df[sum_columns].sum(axis=1) / df[field_names.TOTAL_POP_FIELD]
)
# Calculate college attendance and adjust low income
df[self.COLLEGE_ATTENDANCE_FIELD] = (
df[self.COLLEGE_ATTENDANCE_MALE_ENROLLED_PUBLIC]
+ df[self.COLLEGE_ATTENDANCE_MALE_ENROLLED_PRIVATE]
@ -361,26 +589,75 @@ class CensusACSETL(ExtractTransformLoad):
1 - df[self.COLLEGE_ATTENDANCE_FIELD]
)
# strip columns
df = df[self.COLUMNS_TO_KEEP]
# Save results to self.
self.df = df
# rename columns to be used in score
rename_fields = {
"Percent of individuals < 200% Federal Poverty Line": field_names.POVERTY_LESS_THAN_200_FPL_FIELD,
}
self.df.rename(
columns=rename_fields,
inplace=True,
errors="raise",
# we impute income for both income measures
## TODO: Convert to pydantic for clarity
logger.info("Imputing income information")
ImputeVariables = namedtuple(
"ImputeVariables", ["raw_field_name", "imputed_field_name"]
)
def load(self) -> None:
logger.info("Saving Census ACS Data")
df = calculate_income_measures(
impute_var_named_tup_list=[
ImputeVariables(
raw_field_name=self.POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME,
imputed_field_name=self.IMPUTED_POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME,
),
ImputeVariables(
raw_field_name=self.COLLEGE_ATTENDANCE_FIELD,
imputed_field_name=self.IMPUTED_COLLEGE_ATTENDANCE_FIELD,
),
],
geo_df=df,
geoid_field=self.GEOID_TRACT_FIELD_NAME,
minimum_population_required_for_imputation=self.MINIMUM_POPULATION_REQUIRED_FOR_IMPUTATION,
)
# mkdir census
self.OUTPUT_PATH.mkdir(parents=True, exist_ok=True)
logger.info("Calculating with imputed values")
self.df.to_csv(path_or_buf=self.OUTPUT_PATH / "usa.csv", index=False)
df[
self.ADJUSTED_AND_IMPUTED_POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME
] = (
df[self.POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME].fillna(
df[self.IMPUTED_POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME]
)
- df[self.COLLEGE_ATTENDANCE_FIELD].fillna(
df[self.IMPUTED_COLLEGE_ATTENDANCE_FIELD]
)
# Use clip to ensure that the values are not negative if college attendance
# is very high
).clip(
lower=0
)
# All values should have a value at this point
assert (
# For tracts with >0 population
df[
df[field_names.TOTAL_POP_FIELD]
>= self.MINIMUM_POPULATION_REQUIRED_FOR_IMPUTATION
][
# Then the imputed field should have no nulls
self.ADJUSTED_AND_IMPUTED_POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME
]
.isna()
.sum()
== 0
), "Error: not all values were filled..."
logger.info("Renaming columns...")
df = df.rename(
columns={
self.ADJUSTED_AND_IMPUTED_POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME: field_names.POVERTY_LESS_THAN_200_FPL_IMPUTED_FIELD,
self.POVERTY_LESS_THAN_200_PERCENT_FPL_FIELD_NAME: field_names.POVERTY_LESS_THAN_200_FPL_FIELD,
}
)
# We generate a boolean that is TRUE when there is an imputed income but not a baseline income, and FALSE otherwise.
# This allows us to see which tracts have an imputed income.
df[field_names.IMPUTED_INCOME_FLAG_FIELD_NAME] = (
df[field_names.POVERTY_LESS_THAN_200_FPL_IMPUTED_FIELD].notna()
& df[field_names.POVERTY_LESS_THAN_200_FPL_FIELD].isna()
)
# Save results to self.
self.output_df = df

166
data/data-pipeline/data_pipeline/etl/sources/census_acs/etl_imputations.py Normal file
View file

@ -0,0 +1,166 @@
from typing import Any
from typing import List
from typing import NamedTuple
from typing import Tuple
import geopandas as gpd
import pandas as pd
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
# pylint: disable=unsubscriptable-object
logger = get_module_logger(__name__)
def _get_fips_mask(
geo_df: gpd.GeoDataFrame,
row: gpd.GeoSeries,
fips_digits: int,
geoid_field: str = "GEOID10_TRACT",
) -> pd.Series:
return (
geo_df[geoid_field].str[:fips_digits] == row[geoid_field][:fips_digits]
)
def _get_neighbor_mask(
geo_df: gpd.GeoDataFrame, row: gpd.GeoSeries
) -> pd.Series:
"""Returns neighboring tracts."""
return geo_df["geometry"].touches(row["geometry"])
def _choose_best_mask(
geo_df: gpd.GeoDataFrame,
masks_in_priority_order: List[pd.Series],
column_to_impute: str,
) -> pd.Series:
for mask in masks_in_priority_order:
if any(geo_df[mask][column_to_impute].notna()):
return mask
raise Exception("No mask found")
def _prepare_dataframe_for_imputation(
impute_var_named_tup_list: List[NamedTuple],
geo_df: gpd.GeoDataFrame,
population_field: str,
minimum_population_required_for_imputation: int = 1,
geoid_field: str = "GEOID10_TRACT",
) -> Tuple[Any, gpd.GeoDataFrame]:
"""Helper for imputation.
Given the inputs of `ImputeVariables`, returns list of tracts that need to be
imputed, along with a GeoDataFrame that has a column with the imputed field
"primed", meaning it is a copy of the raw field.
Will drop any rows with population less than
`minimum_population_required_for_imputation`.
"""
imputing_cols = [
impute_var_pair.raw_field_name
for impute_var_pair in impute_var_named_tup_list
]
# Prime column to exist
for impute_var_pair in impute_var_named_tup_list:
geo_df[impute_var_pair.imputed_field_name] = geo_df[
impute_var_pair.raw_field_name
].copy()
# Generate a list of tracts for which at least one of the imputation
# columns is null that also meets population criteria.
tract_list = geo_df[
(
# First, check whether any of the columns we want to impute contain null
# values
geo_df[imputing_cols].isna().any(axis=1)
# Second, ensure population is not null and >= the minimum population
& (
geo_df[population_field].notnull()
& (
geo_df[population_field]
>= minimum_population_required_for_imputation
)
)
)
][geoid_field].unique()
# Check that imputation is a valid choice for this set of fields
logger.info(f"Imputing values for {len(tract_list)} unique tracts.")
assert len(tract_list) > 0, "Error: No missing values to impute"
return tract_list, geo_df
def calculate_income_measures(
impute_var_named_tup_list: list,
geo_df: gpd.GeoDataFrame,
geoid_field: str,
population_field: str = field_names.TOTAL_POP_FIELD,
minimum_population_required_for_imputation: int = 1,
) -> pd.DataFrame:
"""Impute values based on geographic neighbors
We only want to check neighbors a single time, so all variables
that we impute get imputed here.
Takes in:
required:
impute_var_named_tup_list: list of named tuples (imputed field, raw field)
geo_df: geo dataframe that already has the census shapefiles merged
geoid field: tract level ID
Returns: non-geometry pd.DataFrame
"""
# Determine where to impute variables and fill a column with nulls
tract_list, geo_df = _prepare_dataframe_for_imputation(
impute_var_named_tup_list=impute_var_named_tup_list,
geo_df=geo_df,
geoid_field=geoid_field,
population_field=population_field,
minimum_population_required_for_imputation=minimum_population_required_for_imputation,
)
# Iterate through the dataframe to impute in place
## TODO: We should probably convert this to a spatial join now that we are doing >1 imputation and it's taking a lot
## of time, but thinking through how to do this while maintaining the masking will take some time. I think the best
## way would be to (1) spatial join to all neighbors, and then (2) iterate to take the "smallest" set of neighbors...
## but haven't implemented it yet.
for index, row in geo_df.iterrows():
if row[geoid_field] in tract_list:
neighbor_mask = _get_neighbor_mask(geo_df, row)
county_mask = _get_fips_mask(
geo_df=geo_df, row=row, fips_digits=5, geoid_field=geoid_field
)
## TODO: Did CEQ decide to cut this?
state_mask = _get_fips_mask(
geo_df=geo_df, row=row, fips_digits=2, geoid_field=geoid_field
)
# Impute fields for every row missing at least one value using the best possible set of neighbors
# Note that later, we will pull raw.fillna(imputed), so the mechanics of this step aren't critical
for impute_var_pair in impute_var_named_tup_list:
mask_to_use = _choose_best_mask(
geo_df=geo_df,
masks_in_priority_order=[
neighbor_mask,
county_mask,
state_mask,
],
column_to_impute=impute_var_pair.raw_field_name,
)
geo_df.loc[index, impute_var_pair.imputed_field_name] = geo_df[
mask_to_use
][impute_var_pair.raw_field_name].mean()
logger.info("Casting geodataframe as a typical dataframe")
# get rid of the geometry column and cast as a typical df
df = pd.DataFrame(
geo_df[[col for col in geo_df.columns if col != "geometry"]]
)
# finally, return the df
return df

2
data/data-pipeline/data_pipeline/etl/sources/census_acs/etl_utils.py
View file

@ -1,9 +1,9 @@
import os
from pathlib import Path
from typing import List
import censusdata
import pandas as pd
from data_pipeline.etl.sources.census.etl_utils import get_state_fips_codes
from data_pipeline.utils import get_module_logger

3
data/data-pipeline/data_pipeline/etl/sources/census_acs_2010/etl.py
View file

@ -1,11 +1,10 @@
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.sources.census_acs.etl_utils import (
retrieve_census_acs_data,
)
from data_pipeline.utils import get_module_logger
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)

21
data/data-pipeline/data_pipeline/etl/sources/census_acs_median_income/etl.py
View file

@ -1,13 +1,14 @@
import json
from pathlib import Path
import numpy as np
import pandas as pd
import requests
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
from data_pipeline.utils import unzip_file_from_url, download_file_from_url
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import download_file_from_url
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import unzip_file_from_url
logger = get_module_logger(__name__)
@ -282,12 +283,20 @@ class CensusACSMedianIncomeETL(ExtractTransformLoad):
# Download MSA median incomes
logger.info("Starting download of MSA median incomes.")
download = requests.get(self.MSA_MEDIAN_INCOME_URL, verify=None)
download = requests.get(
self.MSA_MEDIAN_INCOME_URL,
verify=None,
timeout=settings.REQUESTS_DEFAULT_TIMOUT,
)
self.msa_median_incomes = json.loads(download.content)
# Download state median incomes
logger.info("Starting download of state median incomes.")
download_state = requests.get(self.STATE_MEDIAN_INCOME_URL, verify=None)
download_state = requests.get(
self.STATE_MEDIAN_INCOME_URL,
verify=None,
timeout=settings.REQUESTS_DEFAULT_TIMOUT,
)
self.state_median_incomes = json.loads(download_state.content)
## NOTE we already have PR's MI here

131
data/data-pipeline/data_pipeline/etl/sources/census_decennial/etl.py
View file

@ -1,12 +1,13 @@
import json
import requests
from typing import List
import numpy as np
import pandas as pd
import requests
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
pd.options.mode.chained_assignment = "raise"
@ -146,6 +147,63 @@ class CensusDecennialETL(ExtractTransformLoad):
field_names.CENSUS_DECENNIAL_UNEMPLOYMENT_FIELD_2009
)
# Race/Ethnicity fields
self.TOTAL_RACE_POPULATION_FIELD = "PCT086001" # Total
self.ASIAN_FIELD = "PCT086002" # Total!!Asian
self.BLACK_FIELD = "PCT086003" # Total!!Black or African American
self.HAWAIIAN_FIELD = (
"PCT086004" # Total!!Native Hawaiian and Other Pacific Islander
)
# Note that the 2010 census for island araeas does not break out
# hispanic and non-hispanic white, so this is slightly different from
# our other demographic data
self.NON_HISPANIC_WHITE_FIELD = "PCT086005" # Total!!White
self.HISPANIC_FIELD = "PCT086006" # Total!!Hispanic or Latino
self.OTHER_RACE_FIELD = "PCT086007" # Total!!Other Ethnic Origin or Ra
self.TOTAL_RACE_POPULATION_VI_FIELD = "P003001" # Total
self.BLACK_VI_FIELD = (
"P003003" # Total!!One race!!Black or African American alone
)
self.AMERICAN_INDIAN_VI_FIELD = "P003005" # Total!!One race!!American Indian and Alaska Native alone
self.ASIAN_VI_FIELD = "P003006" # Total!!One race!!Asian alone
self.HAWAIIAN_VI_FIELD = "P003007" # Total!!One race!!Native Hawaiian and Other Pacific Islander alone
self.TWO_OR_MORE_RACES_VI_FIELD = "P003009" # Total!!Two or More Races
self.NON_HISPANIC_WHITE_VI_FIELD = (
"P005006" # Total!!Not Hispanic or Latino!!One race!!White alone
)
self.HISPANIC_VI_FIELD = "P005002" # Total!!Hispanic or Latino
self.OTHER_RACE_VI_FIELD = (
"P003008" # Total!!One race!!Some Other Race alone
)
self.TOTAL_RACE_POPULATION_VI_FIELD = "P003001" # Total
self.TOTAL_RACE_POPULATION_FIELD_NAME = (
"Total population surveyed on racial data"
)
self.BLACK_FIELD_NAME = "Black or African American"
self.AMERICAN_INDIAN_FIELD_NAME = "American Indian / Alaska Native"
self.ASIAN_FIELD_NAME = "Asian"
self.HAWAIIAN_FIELD_NAME = "Native Hawaiian or Pacific"
self.TWO_OR_MORE_RACES_FIELD_NAME = "two or more races"
self.NON_HISPANIC_WHITE_FIELD_NAME = "White"
self.HISPANIC_FIELD_NAME = "Hispanic or Latino"
# Note that `other` is lowercase because the whole field will show up in the download
# file as "Percent other races"
self.OTHER_RACE_FIELD_NAME = "other races"
# Name output demographics fields.
self.RE_OUTPUT_FIELDS = [
self.BLACK_FIELD_NAME,
self.AMERICAN_INDIAN_FIELD_NAME,
self.ASIAN_FIELD_NAME,
self.HAWAIIAN_FIELD_NAME,
self.TWO_OR_MORE_RACES_FIELD_NAME,
self.NON_HISPANIC_WHITE_FIELD_NAME,
self.HISPANIC_FIELD_NAME,
self.OTHER_RACE_FIELD_NAME,
]
var_list = [
self.MEDIAN_INCOME_FIELD,
self.TOTAL_HOUSEHOLD_RATIO_INCOME_TO_POVERTY_LEVEL_FIELD,
@ -161,6 +219,13 @@ class CensusDecennialETL(ExtractTransformLoad):
self.EMPLOYMENT_FEMALE_IN_LABOR_FORCE_FIELD,
self.EMPLOYMENT_FEMALE_UNEMPLOYED_FIELD,
self.TOTAL_POP_FIELD,
self.TOTAL_RACE_POPULATION_FIELD,
self.ASIAN_FIELD,
self.BLACK_FIELD,
self.HAWAIIAN_FIELD,
self.NON_HISPANIC_WHITE_FIELD,
self.HISPANIC_FIELD,
self.OTHER_RACE_FIELD,
]
var_list = ",".join(var_list)
@ -179,6 +244,15 @@ class CensusDecennialETL(ExtractTransformLoad):
self.EMPLOYMENT_FEMALE_IN_LABOR_FORCE_VI_FIELD,
self.EMPLOYMENT_FEMALE_UNEMPLOYED_VI_FIELD,
self.TOTAL_POP_VI_FIELD,
self.BLACK_VI_FIELD,
self.AMERICAN_INDIAN_VI_FIELD,
self.ASIAN_VI_FIELD,
self.HAWAIIAN_VI_FIELD,
self.TWO_OR_MORE_RACES_VI_FIELD,
self.NON_HISPANIC_WHITE_VI_FIELD,
self.HISPANIC_VI_FIELD,
self.OTHER_RACE_VI_FIELD,
self.TOTAL_RACE_POPULATION_VI_FIELD,
]
var_list_vi = ",".join(var_list_vi)
@ -209,6 +283,23 @@ class CensusDecennialETL(ExtractTransformLoad):
self.EMPLOYMENT_MALE_UNEMPLOYED_FIELD: self.EMPLOYMENT_MALE_UNEMPLOYED_FIELD,
self.EMPLOYMENT_FEMALE_IN_LABOR_FORCE_FIELD: self.EMPLOYMENT_FEMALE_IN_LABOR_FORCE_FIELD,
self.EMPLOYMENT_FEMALE_UNEMPLOYED_FIELD: self.EMPLOYMENT_FEMALE_UNEMPLOYED_FIELD,
self.TOTAL_RACE_POPULATION_FIELD: self.TOTAL_RACE_POPULATION_FIELD_NAME,
self.TOTAL_RACE_POPULATION_VI_FIELD: self.TOTAL_RACE_POPULATION_FIELD_NAME,
# Note there is no American Indian data for AS/GU/MI
self.AMERICAN_INDIAN_VI_FIELD: self.AMERICAN_INDIAN_FIELD_NAME,
self.ASIAN_FIELD: self.ASIAN_FIELD_NAME,
self.ASIAN_VI_FIELD: self.ASIAN_FIELD_NAME,
self.BLACK_FIELD: self.BLACK_FIELD_NAME,
self.BLACK_VI_FIELD: self.BLACK_FIELD_NAME,
self.HAWAIIAN_FIELD: self.HAWAIIAN_FIELD_NAME,
self.HAWAIIAN_VI_FIELD: self.HAWAIIAN_FIELD_NAME,
self.TWO_OR_MORE_RACES_VI_FIELD: self.TWO_OR_MORE_RACES_FIELD_NAME,
self.NON_HISPANIC_WHITE_FIELD: self.NON_HISPANIC_WHITE_FIELD_NAME,
self.NON_HISPANIC_WHITE_VI_FIELD: self.NON_HISPANIC_WHITE_FIELD_NAME,
self.HISPANIC_FIELD: self.HISPANIC_FIELD_NAME,
self.HISPANIC_VI_FIELD: self.HISPANIC_FIELD_NAME,
self.OTHER_RACE_FIELD: self.OTHER_RACE_FIELD_NAME,
self.OTHER_RACE_VI_FIELD: self.OTHER_RACE_FIELD_NAME,
}
# To do: Ask Census Slack Group about whether you need to hardcode the county fips
@ -251,6 +342,8 @@ class CensusDecennialETL(ExtractTransformLoad):
+ "&for=tract:*&in=state:{}%20county:{}"
)
self.final_race_fields: List[str] = []
self.df: pd.DataFrame
self.df_vi: pd.DataFrame
self.df_all: pd.DataFrame
@ -263,14 +356,17 @@ class CensusDecennialETL(ExtractTransformLoad):
f"Downloading data for state/territory {island['state_abbreviation']}"
)
for county in island["county_fips"]:
api_url = self.API_URL.format(
self.DECENNIAL_YEAR,
island["state_abbreviation"],
island["var_list"],
island["fips"],
county,
)
logger.debug(f"CENSUS: Requesting {api_url}")
download = requests.get(
self.API_URL.format(
self.DECENNIAL_YEAR,
island["state_abbreviation"],
island["var_list"],
island["fips"],
county,
)
api_url,
timeout=settings.REQUESTS_DEFAULT_TIMOUT,
)
df = json.loads(download.content)
@ -377,6 +473,19 @@ class CensusDecennialETL(ExtractTransformLoad):
self.df_all["state"] + self.df_all["county"] + self.df_all["tract"]
)
# Calculate stats by race
for race_field_name in self.RE_OUTPUT_FIELDS:
output_field_name = (
field_names.PERCENT_PREFIX
+ race_field_name
+ field_names.ISLAND_AREA_BACKFILL_SUFFIX
)
self.final_race_fields.append(output_field_name)
self.df_all[output_field_name] = (
self.df_all[race_field_name]
/ self.df_all[self.TOTAL_RACE_POPULATION_FIELD_NAME]
)
# Reporting Missing Values
for col in self.df_all.columns:
missing_value_count = self.df_all[col].isnull().sum()
@ -400,7 +509,7 @@ class CensusDecennialETL(ExtractTransformLoad):
self.PERCENTAGE_HOUSEHOLDS_BELOW_200_PERC_POVERTY_LEVEL_FIELD_NAME,
self.PERCENTAGE_HIGH_SCHOOL_ED_FIELD_NAME,
self.UNEMPLOYMENT_FIELD_NAME,
]
] + self.final_race_fields
self.df_all[columns_to_include].to_csv(
path_or_buf=self.OUTPUT_PATH / "usa.csv", index=False

89
data/data-pipeline/data_pipeline/etl/sources/child_opportunity_index/etl.py
View file

@ -1,9 +1,10 @@
from pathlib import Path
import pandas as pd
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger, unzip_file_from_url
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
logger = get_module_logger(__name__)
@ -21,14 +22,35 @@ class ChildOpportunityIndex(ExtractTransformLoad):
Full technical documents: https://www.diversitydatakids.org/sites/default/files/2020-02/ddk_coi2.0_technical_documentation_20200212.pdf.
Github repo: https://github.com/diversitydatakids/COI/
"""
# Metadata for the baseclass
NAME = "child_opportunity_index"
GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
LOAD_YAML_CONFIG: bool = True
# Define these for easy code completion
EXTREME_HEAT_FIELD: str
HEALTHY_FOOD_FIELD: str
IMPENETRABLE_SURFACES_FIELD: str
READING_FIELD: str
PUERTO_RICO_EXPECTED_IN_DATA = False
def __init__(self):
self.COI_FILE_URL = (
"https://data.diversitydatakids.org/datastore/zip/f16fff12-b1e5-4f60-85d3-"
"3a0ededa30a0?format=csv"
)
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
self.SOURCE_URL = (
f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/"
"child_opportunity_index/raw.zip"
)
else:
self.SOURCE_URL = (
"https://data.diversitydatakids.org/datastore/zip/f16fff12-b1e5-4f60-85d3-"
"3a0ededa30a0?format=csv"
)
# TODO: Decide about nixing this
self.TRACT_INPUT_COLUMN_NAME = self.INPUT_GEOID_TRACT_FIELD_NAME
self.OUTPUT_PATH: Path = (
self.DATA_PATH / "dataset" / "child_opportunity_index"
@ -40,31 +62,19 @@ class ChildOpportunityIndex(ExtractTransformLoad):
self.IMPENETRABLE_SURFACES_INPUT_FIELD = "HE_GREEN"
self.READING_INPUT_FIELD = "ED_READING"
# Constants for output
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
field_names.EXTREME_HEAT_FIELD,
field_names.HEALTHY_FOOD_FIELD,
field_names.IMPENETRABLE_SURFACES_FIELD,
field_names.READING_FIELD,
]
self.raw_df: pd.DataFrame
self.output_df: pd.DataFrame
def extract(self) -> None:
logger.info("Starting 51MB data download.")
unzip_file_from_url(
file_url=self.COI_FILE_URL,
download_path=self.get_tmp_path(),
unzipped_file_path=self.get_tmp_path() / "child_opportunity_index",
super().extract(
source_url=self.SOURCE_URL,
extract_path=self.get_tmp_path(),
)
self.raw_df = pd.read_csv(
filepath_or_buffer=self.get_tmp_path()
/ "child_opportunity_index"
/ "raw.csv",
def transform(self) -> None:
logger.info("Starting transforms.")
raw_df = pd.read_csv(
filepath_or_buffer=self.get_tmp_path() / "raw.csv",
# The following need to remain as strings for all of their digits, not get
# converted to numbers.
dtype={
@ -73,16 +83,13 @@ class ChildOpportunityIndex(ExtractTransformLoad):
low_memory=False,
)
def transform(self) -> None:
logger.info("Starting transforms.")
output_df = self.raw_df.rename(
output_df = raw_df.rename(
columns={
self.TRACT_INPUT_COLUMN_NAME: self.GEOID_TRACT_FIELD_NAME,
self.EXTREME_HEAT_INPUT_FIELD: field_names.EXTREME_HEAT_FIELD,
self.HEALTHY_FOOD_INPUT_FIELD: field_names.HEALTHY_FOOD_FIELD,
self.IMPENETRABLE_SURFACES_INPUT_FIELD: field_names.IMPENETRABLE_SURFACES_FIELD,
self.READING_INPUT_FIELD: field_names.READING_FIELD,
self.EXTREME_HEAT_INPUT_FIELD: self.EXTREME_HEAT_FIELD,
self.HEALTHY_FOOD_INPUT_FIELD: self.HEALTHY_FOOD_FIELD,
self.IMPENETRABLE_SURFACES_INPUT_FIELD: self.IMPENETRABLE_SURFACES_FIELD,
self.READING_INPUT_FIELD: self.READING_FIELD,
}
)
@ -95,8 +102,8 @@ class ChildOpportunityIndex(ExtractTransformLoad):
# Convert percents from 0-100 to 0-1 to standardize with our other fields.
percent_fields_to_convert = [
field_names.HEALTHY_FOOD_FIELD,
field_names.IMPENETRABLE_SURFACES_FIELD,
self.HEALTHY_FOOD_FIELD,
self.IMPENETRABLE_SURFACES_FIELD,
]
for percent_field_to_convert in percent_fields_to_convert:
@ -105,11 +112,3 @@ class ChildOpportunityIndex(ExtractTransformLoad):
)
self.output_df = output_df
def load(self) -> None:
logger.info("Saving CSV")
self.OUTPUT_PATH.mkdir(parents=True, exist_ok=True)
self.output_df[self.COLUMNS_TO_KEEP].to_csv(
path_or_buf=self.OUTPUT_PATH / "usa.csv", index=False
)

61
data/data-pipeline/data_pipeline/etl/sources/doe_energy_burden/etl.py
View file

@ -1,64 +1,51 @@
from pathlib import Path
import pandas as pd
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger, unzip_file_from_url
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)
class DOEEnergyBurden(ExtractTransformLoad):
def __init__(self):
self.DOE_FILE_URL = (
settings.AWS_JUSTICE40_DATASOURCES_URL
+ "/DOE_LEAD_AMI_TRACT_2018_ALL.csv.zip"
)
NAME = "doe_energy_burden"
SOURCE_URL: str = (
settings.AWS_JUSTICE40_DATASOURCES_URL
+ "/DOE_LEAD_AMI_TRACT_2018_ALL.csv.zip"
)
GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
LOAD_YAML_CONFIG: bool = True
REVISED_ENERGY_BURDEN_FIELD_NAME: str
def __init__(self):
self.OUTPUT_PATH: Path = (
self.DATA_PATH / "dataset" / "doe_energy_burden"
)
self.TRACT_INPUT_COLUMN_NAME = "FIP"
self.INPUT_ENERGY_BURDEN_FIELD_NAME = "BURDEN"
self.REVISED_ENERGY_BURDEN_FIELD_NAME = "Energy burden"
# Constants for output
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
self.REVISED_ENERGY_BURDEN_FIELD_NAME,
]
self.raw_df: pd.DataFrame
self.output_df: pd.DataFrame
def extract(self) -> None:
logger.info("Starting data download.")
unzip_file_from_url(
file_url=self.DOE_FILE_URL,
download_path=self.get_tmp_path(),
unzipped_file_path=self.get_tmp_path() / "doe_energy_burden",
)
self.raw_df = pd.read_csv(
def transform(self) -> None:
logger.info("Starting DOE Energy Burden transforms.")
raw_df: pd.DataFrame = pd.read_csv(
filepath_or_buffer=self.get_tmp_path()
/ "doe_energy_burden"
/ "DOE_LEAD_AMI_TRACT_2018_ALL.csv",
# The following need to remain as strings for all of their digits, not get converted to numbers.
dtype={
self.TRACT_INPUT_COLUMN_NAME: "string",
self.INPUT_GEOID_TRACT_FIELD_NAME: "string",
},
low_memory=False,
)
def transform(self) -> None:
logger.info("Starting transforms.")
output_df = self.raw_df.rename(
logger.info("Renaming columns and ensuring output format is correct")
output_df = raw_df.rename(
columns={
self.INPUT_ENERGY_BURDEN_FIELD_NAME: self.REVISED_ENERGY_BURDEN_FIELD_NAME,
self.TRACT_INPUT_COLUMN_NAME: self.GEOID_TRACT_FIELD_NAME,
self.INPUT_GEOID_TRACT_FIELD_NAME: self.GEOID_TRACT_FIELD_NAME,
}
)
@ -71,11 +58,3 @@ class DOEEnergyBurden(ExtractTransformLoad):
)
self.output_df = output_df
def load(self) -> None:
logger.info("Saving DOE Energy Burden CSV")
self.OUTPUT_PATH.mkdir(parents=True, exist_ok=True)
self.output_df[self.COLUMNS_TO_KEEP].to_csv(
path_or_buf=self.OUTPUT_PATH / "usa.csv", index=False
)

16
data/data-pipeline/data_pipeline/etl/sources/dot_travel_composite/README.md Normal file
View file

@ -0,0 +1,16 @@
# DOT travel barriers
The below description is taken from DOT directly:
Consistent with OMB’s Interim Guidance for the Justice40 Initiative, DOT’s interim definition of DACs includes (a) certain qualifying census tracts, (b) any Tribal land, or (c) any territory or possession of the United States. DOT has provided a mapping tool to assist applicants in identifying whether a project is located in a Disadvantaged Community, available at Transportation Disadvantaged Census Tracts (arcgis.com). A shapefile of the geospatial data is available Transportation Disadvantaged Census Tracts shapefile (version 2 .0, posted 5/10/22).
The DOT interim definition for DACs was developed by an internal and external collaborative research process (see recordings from November 2021 public meetings). It includes data for 22 indicators collected at the census tract level and grouped into six (6) categories of transportation disadvantage. The numbers in parenthesis show how many indicators fall in that category:
- Transportation access disadvantage identifies communities and places that spend more, and take longer, to get where they need to go. (4)
- Health disadvantage identifies communities based on variables associated with adverse health outcomes, disability, as well as environmental exposures. (3)
- Environmental disadvantage identifies communities with disproportionately high levels of certain air pollutants and high potential presence of lead-based paint in housing units. (6)
- Economic disadvantage identifies areas and populations with high poverty, low wealth, lack of local jobs, low homeownership, low educational attainment, and high inequality. (7)
Resilience disadvantage identifies communities vulnerable to hazards caused by climate change. (1)
- Equity disadvantage identifies communities with a with a high percentile of persons (age 5+) who speak English "less than well." (1)
The CEJST uses only Transportation Access Disadvantage.

0
data/data-pipeline/data_pipeline/etl/sources/dot_travel_composite/__init__.py Normal file
View file

69
data/data-pipeline/data_pipeline/etl/sources/dot_travel_composite/etl.py Normal file
View file

@ -0,0 +1,69 @@
# pylint: disable=unsubscriptable-object
# pylint: disable=unsupported-assignment-operation
import geopandas as gpd
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
logger = get_module_logger(__name__)
class TravelCompositeETL(ExtractTransformLoad):
"""ETL class for the DOT Travel Disadvantage Dataset"""
NAME = "travel_composite"
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
SOURCE_URL = (
f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/"
"dot_travel_composite/Shapefile_and_Metadata.zip"
)
else:
SOURCE_URL = "https://www.transportation.gov/sites/dot.gov/files/Shapefile_and_Metadata.zip"
GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
PUERTO_RICO_EXPECTED_IN_DATA = False
LOAD_YAML_CONFIG: bool = True
# Output score variables (values set on datasets.yml) for linting purposes
TRAVEL_BURDEN_FIELD_NAME: str
def __init__(self):
# define the full path for the input CSV file
self.INPUT_SHP = (
self.get_tmp_path() / "DOT_Disadvantage_Layer_Final_April2022.shp"
)
# this is the main dataframe
self.df: pd.DataFrame
# Start dataset-specific vars here
## Average of Transportation Indicator Percentiles (calculated)
## Calculated: Average of (EPL_TCB+EPL_NWKI+EPL_NOVEH+EPL_COMMUTE) excluding NULLS
## See metadata for more information
self.INPUT_TRAVEL_DISADVANTAGE_FIELD_NAME = "Transp_TH"
self.INPUT_GEOID_TRACT_FIELD_NAME = "FIPS"
def transform(self) -> None:
"""Reads the unzipped data file into memory and applies the following
transformations to prepare it for the load() method:
- Renames the Census Tract column to match the other datasets
- Converts to CSV
"""
logger.info("Transforming DOT Travel Disadvantage Data")
# read in the unzipped shapefile from data source
# reformat it to be standard df, remove unassigned rows, and
# then rename the Census Tract column for merging
df_dot: pd.DataFrame = gpd.read_file(self.INPUT_SHP)
df_dot = df_dot.rename(
columns={
self.INPUT_GEOID_TRACT_FIELD_NAME: self.GEOID_TRACT_FIELD_NAME,
self.INPUT_TRAVEL_DISADVANTAGE_FIELD_NAME: self.TRAVEL_BURDEN_FIELD_NAME,
}
).dropna(subset=[self.GEOID_TRACT_FIELD_NAME])
# Assign the final df to the class' output_df for the load method
self.output_df = df_dot

40
data/data-pipeline/data_pipeline/etl/sources/eamlis/README.md Normal file
View file

@ -0,0 +1,40 @@
The following is the description from eAMLIS as of August 16, 2022.
---
e-AMLIS is not a comprehensive database of all AML features or all AML grant activities. e-AMLIS is a national inventory that provides information about known abandoned mine land (AML) features including polluted waters. The majority of the data in e-AMLIS provides information about known coal AML features for the 25 states and 3 tribal SMCRA-approved AML Programs. e-AMLIS also provides limited information on non-coal AML features, and, non-coal reclamation projects as well as AML features for states and tribes that do not have an approved AML Program. Additionally, e-AMLIS only accounts for the direct construction cost to reclaim each AML feature that has been identified by states and Tribes. Other project costs such as planning, design, permitting, and construction oversight are not tracked in e-AMLIS.
The figures in e-AMLIS are further broken down into 3 cost categories:
Unfunded Cost represents pre-construction estimates to reclaim the AML feature;
Funded Cost indicates that construction has been approved by OSM and these figures may change during construction;
Completed Cost is the actual cost to complete construction and reclamation of the AML feature.
DOI/OSMRE’s Financial Business & Management System is the system of record to obtain comprehensive information about all AML grant expenditures.
An inventory of land and water impacted by past mining (primarily coal mining) is maintained by OSMRE to provide information needed to implement the Surface Mining Control and Reclamation Act of 1977 (SMCRA). The inventory contains information on the location, type, and extent of AML impacts, as well as, information on the cost associated with the reclamation of those problems. The inventory is based upon field surveys by State, Tribal, and OSMRE program officials. It is dynamic to the extent that it is modified as new problems are identified and existing problems are reclaimed.
The Abandoned Mine Land Reclamation Act (AMRA) of 1990, amended SMCRA. The amended law expanded the scope of data OSMRE must collect regarding AML reclamation programs and progress. On December 20, 2006, SMCRA was amended under the Tax Relief and Health Care Act of 2006 to add sources of program funding, emphasize high priority coal reclamation, and expand OSMRE’s responsibilities towards implementation and management of the AML Inventory.
WHO MAINTAINS THE INFORMATION IN THE AML INVENTORY?
The information is developed and/or updated by the States and Indian Tribes managing their own AML programs under SMCRA or by the OSMRE office responsible for States and Indian Tribes not managing their own AML problems.
TYPES OF PROBLEMS
"High Priority"
The most serious AML problems are those posing a threat to health, safety and general welfare of people (Priority 1 and Priority 2, or "high priority"). These are the only problems which the law requires to be inventoried. There are 17 Priority 1 and 2 problem types.
Emergencies
Under the 2006 amendments to SMCRA, AML grants to states and tribes increased from $145 million in FY 2007 to $395 million in FY 2011. The increase in funding allowed states to take responsibility for their AML emergencies as part of their regular AML programs.
Until FY 2011, OSMRE provided Abandoned Mine Land (AML) State Emergency grants to the 15 states that manage their own emergency programs under the Abandoned Mine Land Reclamation Program. Thirteen other states and tribes that had approved AML programs did not receive emergency grants. OSMRE managed emergencies in those 13 states and tribes as well as in Federal Program States without AML programs.
OSMRE officially notified the state and tribal officials and Congressional delegations that, starting on October 1, 2010, they would fully assume responsibility for funding their emergency programs. OSMRE then worked with states and tribes to ensure a smooth transition to the states’ assumption of responsibility for administering state emergency programs. New funding and carryover balances were used during the transition to address immediate needs.
Overall, OSMRE successfully transitioned the financial responsibility to the states in FY 2011, and continues to provide technical and program assistance when needed. States with AML programs are now in a position to effectively handle emergency programs.
Environmental
AML problems impacting the environment are known as Priority 3 problems. While SMCRA does not require OSMRE to inventory every unreclaimed priority 3 problem, some program States and Indian tribes have chosen to submit such information. Information for priority 3 problem types is required when reclamation activities are funded and information on completed reclamation of priority 3 problems is kept in the inventory.
Other Coal Mine Related Problems
Information is also kept on lower priority coal related AML problems such as lower priority coal-related projects involving public facilities, and the development of publicly-owned land. The lower priority problems are also categorized-- Priority 4 and 5 problem types.
Non-coal Mine Related AML Problems
The non-coal problems are primarily problems reclaimed by States/Indian tribes that had "Certified" having addressed all known eligible coal related problems. States and Indian tribes managing their own AML programs reclaimed non-coal problems prior to addressing all their coal related problems under SMCRA SEC. 409-- FILLING VOIDS AND SEALING TUNNELS at the request of the Governor of the state or the governing body of the Indian tribe if the Secretary of the Department of the Interior determines such problems meet the criteria for a priority 1, extreme hazard, problems. This Program Area contains historical reclamation accomplishments for Certified Programs reclaiming Priority 1, 2, and 3 non-coal Problem Type features with pre-AML Reauthorization SMCRA funds distributed prior to October 1, 2007.

0
data/data-pipeline/data_pipeline/etl/sources/eamlis/__init__.py Normal file
View file

81
data/data-pipeline/data_pipeline/etl/sources/eamlis/etl.py Normal file
View file

@ -0,0 +1,81 @@
from pathlib import Path
import geopandas as gpd
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.etl.sources.geo_utils import add_tracts_for_geometries
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)
class AbandonedMineETL(ExtractTransformLoad):
"""Data from Office Of Surface Mining Reclamation and Enforcement's
eAMLIS. These are the locations of abandoned mines.
"""
# Metadata for the baseclass
NAME = "eamlis"
GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
AML_BOOLEAN: str
LOAD_YAML_CONFIG: bool = True
PUERTO_RICO_EXPECTED_IN_DATA = False
EXPECTED_MISSING_STATES = [
"10",
"11",
"12",
"15",
"23",
"27",
"31",
"33",
"34",
"36",
"45",
"50",
"55",
]
# Define these for easy code completion
def __init__(self):
self.SOURCE_URL = (
settings.AWS_JUSTICE40_DATASOURCES_URL
+ "/eAMLIS export of all data.tsv.zip"
)
self.TRACT_INPUT_COLUMN_NAME = self.INPUT_GEOID_TRACT_FIELD_NAME
self.OUTPUT_PATH: Path = (
self.DATA_PATH / "dataset" / "abandoned_mine_land_inventory_system"
)
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
self.AML_BOOLEAN,
]
self.output_df: pd.DataFrame
def transform(self) -> None:
logger.info("Starting eAMLIS transforms.")
df = pd.read_csv(
self.get_tmp_path() / "eAMLIS export of all data.tsv",
sep="\t",
low_memory=False,
)
gdf = gpd.GeoDataFrame(
df,
geometry=gpd.points_from_xy(
x=df["Longitude"],
y=df["Latitude"],
),
crs="epsg:4326",
)
gdf = gdf.drop_duplicates(subset=["geometry"], keep="last")
gdf_tracts = add_tracts_for_geometries(gdf)
gdf_tracts = gdf_tracts.drop_duplicates(self.GEOID_TRACT_FIELD_NAME)
gdf_tracts[self.AML_BOOLEAN] = True
self.output_df = gdf_tracts[self.COLUMNS_TO_KEEP]

39
data/data-pipeline/data_pipeline/etl/sources/ejscreen/etl.py
View file

@ -1,6 +1,6 @@
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
@ -8,21 +8,22 @@ logger = get_module_logger(__name__)
class EJSCREENETL(ExtractTransformLoad):
"""Load EJSCREEN data.
"""Load updated EJSCREEN data."""
Data dictionary:
https://gaftp.epa.gov/EJSCREEN/2019/2019_EJSCREEN_columns_explained.csv
"""
NAME = "ejscreen"
GEO_LEVEL: ValidGeoLevel = ValidGeoLevel.CENSUS_TRACT
INPUT_GEOID_TRACT_FIELD_NAME: str = "ID"
def __init__(self):
self.EJSCREEN_FTP_URL = "https://edap-arcgiscloud-data-commons.s3.amazonaws.com/EJSCREEN2020/EJSCREEN_Tract_2020_USPR.csv.zip"
self.EJSCREEN_CSV = self.get_tmp_path() / "EJSCREEN_Tract_2020_USPR.csv"
self.CSV_PATH = self.DATA_PATH / "dataset" / "ejscreen_2019"
self.EJSCREEN_FTP_URL = "https://gaftp.epa.gov/EJSCREEN/2021/EJSCREEN_2021_USPR_Tracts.csv.zip"
self.EJSCREEN_CSV = (
self.get_tmp_path() / "EJSCREEN_2021_USPR_Tracts.csv"
)
self.CSV_PATH = self.DATA_PATH / "dataset" / "ejscreen"
self.df: pd.DataFrame
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
field_names.TOTAL_POP_FIELD,
# pylint: disable=duplicate-code
field_names.AIR_TOXICS_CANCER_RISK_FIELD,
field_names.RESPIRATORY_HAZARD_FIELD,
@ -39,6 +40,7 @@ class EJSCREENETL(ExtractTransformLoad):
field_names.OVER_64_FIELD,
field_names.UNDER_5_FIELD,
field_names.LEAD_PAINT_FIELD,
field_names.UST_FIELD,
]
def extract(self) -> None:
@ -53,19 +55,16 @@ class EJSCREENETL(ExtractTransformLoad):
logger.info("Transforming EJScreen Data")
self.df = pd.read_csv(
self.EJSCREEN_CSV,
dtype={"ID": "string"},
dtype={self.INPUT_GEOID_TRACT_FIELD_NAME: str},
# EJSCREEN writes the word "None" for NA data.
na_values=["None"],
low_memory=False,
)
# rename ID to Tract ID
self.df.rename(
self.output_df = self.df.rename(
columns={
"ID": self.GEOID_TRACT_FIELD_NAME,
# Note: it is currently unorthodox to use `field_names` in an ETL class,
# but I think that's the direction we'd like to move all ETL classes. - LMB
"ACSTOTPOP": field_names.TOTAL_POP_FIELD,
self.INPUT_GEOID_TRACT_FIELD_NAME: self.GEOID_TRACT_FIELD_NAME,
"CANCER": field_names.AIR_TOXICS_CANCER_RISK_FIELD,
"RESP": field_names.RESPIRATORY_HAZARD_FIELD,
"DSLPM": field_names.DIESEL_FIELD,
@ -81,14 +80,6 @@ class EJSCREENETL(ExtractTransformLoad):
"OVER64PCT": field_names.OVER_64_FIELD,
"UNDER5PCT": field_names.UNDER_5_FIELD,
"PRE1960PCT": field_names.LEAD_PAINT_FIELD,
"UST": field_names.UST_FIELD, # added for 2021 update
},
inplace=True,
)
def load(self) -> None:
logger.info("Saving EJScreen CSV")
# write nationwide csv
self.CSV_PATH.mkdir(parents=True, exist_ok=True)
self.df[self.COLUMNS_TO_KEEP].to_csv(
self.CSV_PATH / "usa.csv", index=False
)

2
data/data-pipeline/data_pipeline/etl/sources/ejscreen_areas_of_concern/etl.py
View file

@ -1,5 +1,4 @@
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger
@ -58,7 +57,6 @@ class EJSCREENAreasOfConcernETL(ExtractTransformLoad):
# TO DO: As a one off we did all the processing in a separate Notebook
# Can add here later for a future PR
pass
def load(self) -> None:
if self.ejscreen_areas_of_concern_data_exists():

5
data/data-pipeline/data_pipeline/etl/sources/energy_definition_alternative_draft/etl.py
View file

@ -1,10 +1,11 @@
from pathlib import Path
import pandas as pd
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger, unzip_file_from_url
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import unzip_file_from_url
logger = get_module_logger(__name__)

18
data/data-pipeline/data_pipeline/etl/sources/epa_rsei/etl.py
View file

@ -1,9 +1,11 @@
from pathlib import Path
import pandas as pd
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger, unzip_file_from_url
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import unzip_file_from_url
from data_pipeline.config import settings
logger = get_module_logger(__name__)
@ -20,7 +22,17 @@ class EPARiskScreeningEnvironmentalIndicatorsETL(ExtractTransformLoad):
"""
def __init__(self):
self.AGGREGATED_RSEI_SCORE_FILE_URL = "http://abt-rsei.s3.amazonaws.com/microdata2019/census_agg/CensusMicroTracts2019_2019_aggregated.zip"
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
self.AGGREGATED_RSEI_SCORE_FILE_URL = (
f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/"
"epa_rsei/CensusMicroTracts2019_2019_aggregated.zip"
)
else:
self.AGGREGATED_RSEI_SCORE_FILE_URL = (
"http://abt-rsei.s3.amazonaws.com/microdata2019/"
"census_agg/CensusMicroTracts2019_2019_aggregated.zip"
)
self.OUTPUT_PATH: Path = self.DATA_PATH / "dataset" / "epa_rsei"
self.EPA_RSEI_SCORE_THRESHOLD_CUTOFF = 0.75

3
data/data-pipeline/data_pipeline/etl/sources/fsf_flood_risk/README.md Normal file
View file

@ -0,0 +1,3 @@
# FSF flood risk data
Flood risk computed as 1 in 100 year flood zone

0
data/data-pipeline/data_pipeline/etl/sources/fsf_flood_risk/__init__.py Normal file
View file

86
data/data-pipeline/data_pipeline/etl/sources/fsf_flood_risk/etl.py Normal file
View file

@ -0,0 +1,86 @@
# pylint: disable=unsubscriptable-object
# pylint: disable=unsupported-assignment-operation
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)
class FloodRiskETL(ExtractTransformLoad):
"""ETL class for the First Street Foundation flood risk dataset"""
NAME = "fsf_flood_risk"
# These data were emailed to the J40 team while first street got
# their official data sharing channels setup.
SOURCE_URL = settings.AWS_JUSTICE40_DATASOURCES_URL + "/fsf_flood.zip"
GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
LOAD_YAML_CONFIG: bool = True
# Output score variables (values set on datasets.yml) for linting purposes
COUNT_PROPERTIES: str
PROPERTIES_AT_RISK_FROM_FLOODING_TODAY: str
PROPERTIES_AT_RISK_FROM_FLOODING_IN_30_YEARS: str
SHARE_OF_PROPERTIES_AT_RISK_FROM_FLOODING_TODAY: str
SHARE_OF_PROPERTIES_AT_RISK_FROM_FLOODING_IN_30_YEARS: str
def __init__(self):
# define the full path for the input CSV file
self.INPUT_CSV = (
self.get_tmp_path() / "fsf_flood" / "flood-tract2010.csv"
)
# this is the main dataframe
self.df: pd.DataFrame
# Start dataset-specific vars here
self.COUNT_PROPERTIES_NATIVE_FIELD_NAME = "count_properties"
self.COUNT_PROPERTIES_AT_RISK_TODAY = "mid_depth_100_year00"
self.COUNT_PROPERTIES_AT_RISK_30_YEARS = "mid_depth_100_year30"
self.CLIP_PROPERTIES_COUNT = 250
def transform(self) -> None:
"""Reads the unzipped data file into memory and applies the following
transformations to prepare it for the load() method:
- Renames the Census Tract column to match the other datasets
- Calculates share of properties at risk, left-clipping number of properties at 250
"""
logger.info("Transforming National Risk Index Data")
# read in the unzipped csv data source then rename the
# Census Tract column for merging
df_fsf_flood: pd.DataFrame = pd.read_csv(
self.INPUT_CSV,
dtype={self.INPUT_GEOID_TRACT_FIELD_NAME: str},
low_memory=False,
)
df_fsf_flood[self.GEOID_TRACT_FIELD_NAME] = df_fsf_flood[
self.INPUT_GEOID_TRACT_FIELD_NAME
].str.zfill(11)
df_fsf_flood[self.COUNT_PROPERTIES] = df_fsf_flood[
self.COUNT_PROPERTIES_NATIVE_FIELD_NAME
].clip(lower=self.CLIP_PROPERTIES_COUNT)
df_fsf_flood[self.SHARE_OF_PROPERTIES_AT_RISK_FROM_FLOODING_TODAY] = (
df_fsf_flood[self.COUNT_PROPERTIES_AT_RISK_TODAY]
/ df_fsf_flood[self.COUNT_PROPERTIES]
)
df_fsf_flood[
self.SHARE_OF_PROPERTIES_AT_RISK_FROM_FLOODING_IN_30_YEARS
] = (
df_fsf_flood[self.COUNT_PROPERTIES_AT_RISK_30_YEARS]
/ df_fsf_flood[self.COUNT_PROPERTIES]
)
# Assign the final df to the class' output_df for the load method with rename
self.output_df = df_fsf_flood.rename(
columns={
self.COUNT_PROPERTIES_AT_RISK_TODAY: self.PROPERTIES_AT_RISK_FROM_FLOODING_TODAY,
self.COUNT_PROPERTIES_AT_RISK_30_YEARS: self.PROPERTIES_AT_RISK_FROM_FLOODING_IN_30_YEARS,
}
)

3
data/data-pipeline/data_pipeline/etl/sources/fsf_wildfire_risk/README.md Normal file
View file

@ -0,0 +1,3 @@
# FSF wildfire risk data
Fire risk computed as >= 0.003 burn risk probability

0
data/data-pipeline/data_pipeline/etl/sources/fsf_wildfire_risk/__init__.py Normal file
View file

83
data/data-pipeline/data_pipeline/etl/sources/fsf_wildfire_risk/etl.py Normal file
View file

@ -0,0 +1,83 @@
# pylint: disable=unsubscriptable-object
# pylint: disable=unsupported-assignment-operation
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)
class WildfireRiskETL(ExtractTransformLoad):
"""ETL class for the First Street Foundation wildfire risk dataset"""
NAME = "fsf_wildfire_risk"
# These data were emailed to the J40 team while first street got
# their official data sharing channels setup.
SOURCE_URL = settings.AWS_JUSTICE40_DATASOURCES_URL + "/fsf_fire.zip"
GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
PUERTO_RICO_EXPECTED_IN_DATA = False
LOAD_YAML_CONFIG: bool = True
ALASKA_AND_HAWAII_EXPECTED_IN_DATA = False
# Output score variables (values set on datasets.yml) for linting purposes
COUNT_PROPERTIES: str
PROPERTIES_AT_RISK_FROM_FIRE_TODAY: str
PROPERTIES_AT_RISK_FROM_FIRE_IN_30_YEARS: str
SHARE_OF_PROPERTIES_AT_RISK_FROM_FIRE_TODAY: str
SHARE_OF_PROPERTIES_AT_RISK_FROM_FIRE_IN_30_YEARS: str
def __init__(self):
# define the full path for the input CSV file
self.INPUT_CSV = self.get_tmp_path() / "fsf_fire" / "fire-tract2010.csv"
# this is the main dataframe
self.df: pd.DataFrame
# Start dataset-specific vars here
self.COUNT_PROPERTIES_NATIVE_FIELD_NAME = "count_properties"
self.COUNT_PROPERTIES_AT_RISK_TODAY = "burnprob_year00_flag"
self.COUNT_PROPERTIES_AT_RISK_30_YEARS = "burnprob_year30_flag"
self.CLIP_PROPERTIES_COUNT = 250
def transform(self) -> None:
"""Reads the unzipped data file into memory and applies the following
transformations to prepare it for the load() method:
- Renames the Census Tract column to match the other datasets
- Calculates share of properties at risk, left-clipping number of properties at 250
"""
logger.info("Transforming National Risk Index Data")
# read in the unzipped csv data source then rename the
# Census Tract column for merging
df_fsf_fire: pd.DataFrame = pd.read_csv(
self.INPUT_CSV,
dtype={self.INPUT_GEOID_TRACT_FIELD_NAME: str},
low_memory=False,
)
df_fsf_fire[self.GEOID_TRACT_FIELD_NAME] = df_fsf_fire[
self.INPUT_GEOID_TRACT_FIELD_NAME
].str.zfill(11)
df_fsf_fire[self.COUNT_PROPERTIES] = df_fsf_fire[
self.COUNT_PROPERTIES_NATIVE_FIELD_NAME
].clip(lower=self.CLIP_PROPERTIES_COUNT)
df_fsf_fire[self.SHARE_OF_PROPERTIES_AT_RISK_FROM_FIRE_TODAY] = (
df_fsf_fire[self.COUNT_PROPERTIES_AT_RISK_TODAY]
/ df_fsf_fire[self.COUNT_PROPERTIES]
)
df_fsf_fire[self.SHARE_OF_PROPERTIES_AT_RISK_FROM_FIRE_IN_30_YEARS] = (
df_fsf_fire[self.COUNT_PROPERTIES_AT_RISK_30_YEARS]
/ df_fsf_fire[self.COUNT_PROPERTIES]
)
# Assign the final df to the class' output_df for the load method with rename
self.output_df = df_fsf_fire.rename(
columns={
self.COUNT_PROPERTIES_AT_RISK_TODAY: self.PROPERTIES_AT_RISK_FROM_FIRE_TODAY,
self.COUNT_PROPERTIES_AT_RISK_30_YEARS: self.PROPERTIES_AT_RISK_FROM_FIRE_IN_30_YEARS,
}
)

92
data/data-pipeline/data_pipeline/etl/sources/geo_utils.py Normal file
View file

@ -0,0 +1,92 @@
"""Utililities for turning geographies into tracts, using census data"""
from functools import lru_cache
from pathlib import Path
from typing import Optional
import geopandas as gpd
from data_pipeline.etl.sources.tribal.etl import TribalETL
from data_pipeline.utils import get_module_logger
from .census.etl import CensusETL
logger = get_module_logger(__name__)
@lru_cache()
def get_tract_geojson(
_tract_data_path: Optional[Path] = None,
) -> gpd.GeoDataFrame:
logger.info("Loading tract geometry data from census ETL")
GEOJSON_PATH = _tract_data_path
if GEOJSON_PATH is None:
GEOJSON_PATH = CensusETL.NATIONAL_TRACT_JSON_PATH
if not GEOJSON_PATH.exists():
logger.debug("Census data has not been computed, running")
census_etl = CensusETL()
census_etl.extract()
census_etl.transform()
census_etl.load()
tract_data = gpd.read_file(
GEOJSON_PATH,
include_fields=["GEOID10"],
)
tract_data = tract_data.rename(
columns={"GEOID10": "GEOID10_TRACT"}, errors="raise"
)
return tract_data
@lru_cache()
def get_tribal_geojson(
_tribal_data_path: Optional[Path] = None,
) -> gpd.GeoDataFrame:
logger.info("Loading Tribal geometry data from Tribal ETL")
GEOJSON_PATH = _tribal_data_path
if GEOJSON_PATH is None:
GEOJSON_PATH = TribalETL().NATIONAL_TRIBAL_GEOJSON_PATH
if not GEOJSON_PATH.exists():
logger.debug("Tribal data has not been computed, running")
tribal_etl = TribalETL()
tribal_etl.extract()
tribal_etl.transform()
tribal_etl.load()
tribal_data = gpd.read_file(
GEOJSON_PATH,
)
return tribal_data
def add_tracts_for_geometries(
df: gpd.GeoDataFrame, tract_data: Optional[gpd.GeoDataFrame] = None
) -> gpd.GeoDataFrame:
"""Adds tract-geoids to dataframe df that contains spatial geometries
Depends on CensusETL for the geodata to do its conversion
Args:
df (GeoDataFrame): a geopandas GeoDataFrame with a point geometry column
tract_data (GeoDataFrame): optional override to directly pass a
geodataframe of the tract boundaries. Also helps simplify testing.
Returns:
GeoDataFrame: the above dataframe, with an additional GEOID10_TRACT column that
maps the points in DF to census tracts and a geometry column for later
spatial analysis
"""
logger.debug("Appending tract data to dataframe")
if tract_data is None:
tract_data = get_tract_geojson()
else:
logger.debug("Using existing tract data.")
assert (
tract_data.crs == df.crs
), f"Dataframe must be projected to {tract_data.crs}"
df = gpd.sjoin(
df,
tract_data[["GEOID10_TRACT", "geometry"]],
how="inner",
op="intersects",
)
return df

38
data/data-pipeline/data_pipeline/etl/sources/geocorr/etl.py
View file

@ -1,16 +1,18 @@
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import (
get_module_logger,
unzip_file_from_url,
)
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import unzip_file_from_url
logger = get_module_logger(__name__)
class GeoCorrETL(ExtractTransformLoad):
NAME = "geocorr"
GEO_LEVEL: ValidGeoLevel = ValidGeoLevel.CENSUS_TRACT
PUERTO_RICO_EXPECTED_IN_DATA = False
def __init__(self):
self.OUTPUT_PATH = self.DATA_PATH / "dataset" / "geocorr"
@ -24,6 +26,10 @@ class GeoCorrETL(ExtractTransformLoad):
self.GEOCORR_PLACES_URL = "https://justice40-data.s3.amazonaws.com/data-sources/geocorr_urban_rural.csv.zip"
self.GEOCORR_GEOID_FIELD_NAME = "GEOID10_TRACT"
self.URBAN_HEURISTIC_FIELD_NAME = "Urban Heuristic Flag"
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
self.URBAN_HEURISTIC_FIELD_NAME,
]
self.df: pd.DataFrame
@ -35,13 +41,11 @@ class GeoCorrETL(ExtractTransformLoad):
file_url=settings.AWS_JUSTICE40_DATASOURCES_URL
+ "/geocorr_urban_rural.csv.zip",
download_path=self.get_tmp_path(),
unzipped_file_path=self.get_tmp_path() / "geocorr",
unzipped_file_path=self.get_tmp_path(),
)
self.df = pd.read_csv(
filepath_or_buffer=self.get_tmp_path()
/ "geocorr"
/ "geocorr_urban_rural.csv",
filepath_or_buffer=self.get_tmp_path() / "geocorr_urban_rural.csv",
dtype={
self.GEOCORR_GEOID_FIELD_NAME: "string",
},
@ -50,22 +54,10 @@ class GeoCorrETL(ExtractTransformLoad):
def transform(self) -> None:
logger.info("Starting GeoCorr Urban Rural Map transform")
# Put in logic from Jupyter Notebook transform when we switch in the hyperlink to Geocorr
self.df.rename(
self.output_df = self.df.rename(
columns={
"urban_heuristic_flag": self.URBAN_HEURISTIC_FIELD_NAME,
},
inplace=True,
)
pass
# Put in logic from Jupyter Notebook transform when we switch in the hyperlink to Geocorr
def load(self) -> None:
logger.info("Saving GeoCorr Urban Rural Map Data")
# mkdir census
self.OUTPUT_PATH.mkdir(parents=True, exist_ok=True)
self.df.to_csv(path_or_buf=self.OUTPUT_PATH / "usa.csv", index=False)

0
data/data-pipeline/data_pipeline/etl/sources/historic_redlining/README.md Normal file
View file

0
data/data-pipeline/data_pipeline/etl/sources/historic_redlining/__init__.py Normal file
View file

70
data/data-pipeline/data_pipeline/etl/sources/historic_redlining/etl.py Normal file
View file

@ -0,0 +1,70 @@
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)
class HistoricRedliningETL(ExtractTransformLoad):
NAME = "historic_redlining"
GEO_LEVEL: ValidGeoLevel = ValidGeoLevel.CENSUS_TRACT
EXPECTED_MISSING_STATES = [
"10",
"11",
"16",
"23",
"30",
"32",
"35",
"38",
"46",
"50",
"56",
]
PUERTO_RICO_EXPECTED_IN_DATA = False
ALASKA_AND_HAWAII_EXPECTED_IN_DATA: bool = False
SOURCE_URL = settings.AWS_JUSTICE40_DATASOURCES_URL + "/HRS_2010.zip"
def __init__(self):
self.CSV_PATH = self.DATA_PATH / "dataset" / "historic_redlining"
self.HISTORIC_REDLINING_FILE_PATH = (
self.get_tmp_path() / "HRS_2010.xlsx"
)
self.REDLINING_SCALAR = "Tract-level redlining score"
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
self.REDLINING_SCALAR,
]
self.df: pd.DataFrame
def transform(self) -> None:
logger.info("Transforming Historic Redlining Data")
# this is obviously temporary
historic_redlining_data = pd.read_excel(
self.HISTORIC_REDLINING_FILE_PATH
)
historic_redlining_data[self.GEOID_TRACT_FIELD_NAME] = (
historic_redlining_data["GEOID10"].astype(str).str.zfill(11)
)
historic_redlining_data = historic_redlining_data.rename(
columns={"HRS2010": self.REDLINING_SCALAR}
)
logger.info(f"{historic_redlining_data.columns}")
# Calculate lots of different score thresholds for convenience
for threshold in [3.25, 3.5, 3.75]:
historic_redlining_data[
f"{self.REDLINING_SCALAR} meets or exceeds {round(threshold, 2)}"
] = (historic_redlining_data[self.REDLINING_SCALAR] >= threshold)
## NOTE We add to columns to keep here
self.COLUMNS_TO_KEEP.append(
f"{self.REDLINING_SCALAR} meets or exceeds {round(threshold, 2)}"
)
self.output_df = historic_redlining_data

8
data/data-pipeline/data_pipeline/etl/sources/housing_and_transportation/etl.py
View file

@ -1,9 +1,9 @@
import pandas as pd
from pandas.errors import EmptyDataError
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.sources.census.etl_utils import get_state_fips_codes
from data_pipeline.utils import get_module_logger, unzip_file_from_url
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import unzip_file_from_url
from pandas.errors import EmptyDataError
logger = get_module_logger(__name__)
@ -35,7 +35,7 @@ class HousingTransportationETL(ExtractTransformLoad):
# New file name:
tmp_csv_file_path = (
zip_file_dir / f"htaindex_data_tracts_{fips}.csv"
zip_file_dir / f"htaindex2019_data_tracts_{fips}.csv"
)
try:

184
data/data-pipeline/data_pipeline/etl/sources/hud_housing/etl.py
View file

@ -1,16 +1,28 @@
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
logger = get_module_logger(__name__)
class HudHousingETL(ExtractTransformLoad):
NAME = "hud_housing"
GEO_LEVEL: ValidGeoLevel = ValidGeoLevel.CENSUS_TRACT
def __init__(self):
self.OUTPUT_PATH = self.DATA_PATH / "dataset" / "hud_housing"
self.GEOID_TRACT_FIELD_NAME = "GEOID10_TRACT"
self.HOUSING_FTP_URL = "https://www.huduser.gov/portal/datasets/cp/2014thru2018-140-csv.zip"
self.HOUSING_ZIP_FILE_DIR = self.get_tmp_path() / "hud_housing"
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
self.HOUSING_FTP_URL = (
f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/"
"hud_housing/2014thru2018-140-csv.zip"
)
else:
self.HOUSING_FTP_URL = "https://www.huduser.gov/portal/datasets/cp/2014thru2018-140-csv.zip"
self.HOUSING_ZIP_FILE_DIR = self.get_tmp_path()
# We measure households earning less than 80% of HUD Area Median Family Income by county
# and paying greater than 30% of their income to housing costs.
@ -19,6 +31,17 @@ class HudHousingETL(ExtractTransformLoad):
self.HOUSING_BURDEN_DENOMINATOR_FIELD_NAME = (
"HOUSING_BURDEN_DENOMINATOR"
)
self.NO_KITCHEN_OR_INDOOR_PLUMBING_FIELD_NAME = (
"Share of homes with no kitchen or indoor plumbing (percent)"
)
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
self.HOUSING_BURDEN_NUMERATOR_FIELD_NAME,
self.HOUSING_BURDEN_DENOMINATOR_FIELD_NAME,
self.HOUSING_BURDEN_FIELD_NAME,
self.NO_KITCHEN_OR_INDOOR_PLUMBING_FIELD_NAME,
"DENOM INCL NOT COMPUTED",
]
# Note: some variable definitions.
# HUD-adjusted median family income (HAMFI).
@ -27,7 +50,8 @@ class HudHousingETL(ExtractTransformLoad):
# - incomplete plumbing facilities,
# - more than 1 person per room,
# - cost burden greater than 30%.
# Table 8 is the desired table.
# Table 8 is the desired table for housing burden
# Table 3 is the desired table for no kitchen or indoor plumbing
self.df: pd.DataFrame
@ -38,124 +62,74 @@ class HudHousingETL(ExtractTransformLoad):
self.HOUSING_ZIP_FILE_DIR,
)
def transform(self) -> None:
logger.info("Transforming HUD Housing Data")
def _read_chas_table(self, file_name):
# New file name:
tmp_csv_file_path = self.HOUSING_ZIP_FILE_DIR / "140" / "Table8.csv"
self.df = pd.read_csv(
tmp_csv_file_path = self.HOUSING_ZIP_FILE_DIR / "140" / file_name
tmp_df = pd.read_csv(
filepath_or_buffer=tmp_csv_file_path,
encoding="latin-1",
)
# Rename and reformat block group ID
self.df.rename(
columns={"geoid": self.GEOID_TRACT_FIELD_NAME}, inplace=True
)
# The CHAS data has census tract ids such as `14000US01001020100`
# Whereas the rest of our data uses, for the same tract, `01001020100`.
# the characters before `US`:
self.df[self.GEOID_TRACT_FIELD_NAME] = self.df[
self.GEOID_TRACT_FIELD_NAME
].str.replace(r"^.*?US", "", regex=True)
# This reformats and renames this field.
tmp_df[self.GEOID_TRACT_FIELD_NAME] = tmp_df["geoid"].str.replace(
r"^.*?US", "", regex=True
)
return tmp_df
def transform(self) -> None:
logger.info("Transforming HUD Housing Data")
table_8 = self._read_chas_table("Table8.csv")
table_3 = self._read_chas_table("Table3.csv")
self.df = table_8.merge(
table_3, how="outer", on=self.GEOID_TRACT_FIELD_NAME
)
# Calculate share that lacks indoor plumbing or kitchen
# This is computed as
# (
# owner occupied without plumbing + renter occupied without plumbing
# ) / (
# total of owner and renter occupied
# )
self.df[self.NO_KITCHEN_OR_INDOOR_PLUMBING_FIELD_NAME] = (
# T3_est3: owner-occupied lacking complete plumbing or kitchen facilities for all levels of income
# T3_est46: subtotal: renter-occupied lacking complete plumbing or kitchen facilities for all levels of income
# T3_est2: subtotal: owner-occupied for all levels of income
# T3_est45: subtotal: renter-occupied for all levels of income
self.df["T3_est3"]
+ self.df["T3_est46"]
) / (self.df["T3_est2"] + self.df["T3_est45"])
# Calculate housing burden
# This is quite a number of steps. It does not appear to be accessible nationally in a simpler format, though.
# See "CHAS data dictionary 12-16.xlsx"
# Owner occupied numerator fields
OWNER_OCCUPIED_NUMERATOR_FIELDS = [
# Column Name
# Line_Type
# Tenure
# Household income
# Cost burden
# Facilities
"T8_est7",
# Subtotal
# Owner occupied
# less than or equal to 30% of HAMFI
# greater than 30% but less than or equal to 50%
# All
"T8_est10",
# Subtotal
# Owner occupied
# less than or equal to 30% of HAMFI
# greater than 50%
# All
"T8_est20",
# Subtotal
# Owner occupied
# greater than 30% but less than or equal to 50% of HAMFI
# greater than 30% but less than or equal to 50%
# All
"T8_est23",
# Subtotal
# Owner occupied
# greater than 30% but less than or equal to 50% of HAMFI
# greater than 50%
# All
"T8_est33",
# Subtotal
# Owner occupied
# greater than 50% but less than or equal to 80% of HAMFI
# greater than 30% but less than or equal to 50%
# All
"T8_est36",
# Subtotal
# Owner occupied
# greater than 50% but less than or equal to 80% of HAMFI
# greater than 50%
# All
"T8_est7", # Owner, less than or equal to 30% of HAMFI, greater than 30% but less than or equal to 50%
"T8_est10", # Owner, less than or equal to 30% of HAMFI, greater than 50%
"T8_est20", # Owner, greater than 30% but less than or equal to 50% of HAMFI, greater than 30% but less than or equal to 50%
"T8_est23", # Owner, greater than 30% but less than or equal to 50% of HAMFI, greater than 50%
"T8_est33", # Owner, greater than 50% but less than or equal to 80% of HAMFI, greater than 30% but less than or equal to 50%
"T8_est36", # Owner, greater than 50% but less than or equal to 80% of HAMFI, greater than 50%
]
# These rows have the values where HAMFI was not computed, b/c of no or negative income.
# They are in the same order as the rows above
OWNER_OCCUPIED_NOT_COMPUTED_FIELDS = [
# Column Name
# Line_Type
# Tenure
# Household income
# Cost burden
# Facilities
"T8_est13",
# Subtotal
# Owner occupied
# less than or equal to 30% of HAMFI
# not computed (no/negative income)
# All
"T8_est26",
# Subtotal
# Owner occupied
# greater than 30% but less than or equal to 50% of HAMFI
# not computed (no/negative income)
# All
"T8_est39",
# Subtotal
# Owner occupied
# greater than 50% but less than or equal to 80% of HAMFI
# not computed (no/negative income)
# All
"T8_est52",
# Subtotal
# Owner occupied
# greater than 80% but less than or equal to 100% of HAMFI
# not computed (no/negative income)
# All
"T8_est65",
# Subtotal
# Owner occupied
# greater than 100% of HAMFI
# not computed (no/negative income)
# All
]
# This represents all owner-occupied housing units
OWNER_OCCUPIED_POPULATION_FIELD = "T8_est2"
# Subtotal
# Owner occupied
# All
# All
# All
# Renter occupied numerator fields
RENTER_OCCUPIED_NUMERATOR_FIELDS = [
@ -280,18 +254,4 @@ class HudHousingETL(ExtractTransformLoad):
float
)
def load(self) -> None:
logger.info("Saving HUD Housing Data")
self.OUTPUT_PATH.mkdir(parents=True, exist_ok=True)
# Drop unnecessary fields
self.df[
[
self.GEOID_TRACT_FIELD_NAME,
self.HOUSING_BURDEN_NUMERATOR_FIELD_NAME,
self.HOUSING_BURDEN_DENOMINATOR_FIELD_NAME,
self.HOUSING_BURDEN_FIELD_NAME,
"DENOM INCL NOT COMPUTED",
]
].to_csv(path_or_buf=self.OUTPUT_PATH / "usa.csv", index=False)
self.output_df = self.df

23
data/data-pipeline/data_pipeline/etl/sources/hud_recap/etl.py
View file

@ -1,16 +1,27 @@
import pandas as pd
import requests
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)
class HudRecapETL(ExtractTransformLoad):
def __init__(self):
# pylint: disable=line-too-long
self.HUD_RECAP_CSV_URL = "https://opendata.arcgis.com/api/v3/datasets/56de4edea8264fe5a344da9811ef5d6e_0/downloads/data?format=csv&spatialRefId=4326" # noqa: E501
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
self.HUD_RECAP_CSV_URL = (
f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/"
"hud_recap/Racially_or_Ethnically_Concentrated_Areas_of_Poverty__R_ECAPs_.csv"
)
else:
self.HUD_RECAP_CSV_URL = (
"https://opendata.arcgis.com/api/v3/datasets/"
"56de4edea8264fe5a344da9811ef5d6e_0/downloads/data?format=csv&spatialRefId=4326"
)
self.HUD_RECAP_CSV = (
self.get_tmp_path()
/ "Racially_or_Ethnically_Concentrated_Areas_of_Poverty__R_ECAPs_.csv"
@ -26,7 +37,11 @@ class HudRecapETL(ExtractTransformLoad):
def extract(self) -> None:
logger.info("Downloading HUD Recap Data")
download = requests.get(self.HUD_RECAP_CSV_URL, verify=None)
download = requests.get(
self.HUD_RECAP_CSV_URL,
verify=None,
timeout=settings.REQUESTS_DEFAULT_TIMOUT,
)
file_contents = download.content
csv_file = open(self.HUD_RECAP_CSV, "wb")
csv_file.write(file_contents)

10
data/data-pipeline/data_pipeline/etl/sources/mapping_for_ej/etl.py
View file

@ -1,10 +1,9 @@
import pandas as pd
import geopandas as gpd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger
from data_pipeline.score import field_names
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)
@ -96,4 +95,3 @@ class MappingForEJETL(ExtractTransformLoad):
def validate(self) -> None:
logger.info("Validating Mapping For EJ Data")
pass

2
data/data-pipeline/data_pipeline/etl/sources/mapping_inequality/data/holc_grades_manually_mapped.csv
View file

@ -37,4 +37,4 @@ Oklahoma City,90R,D
Milwaukee Co.,S-D1,D
Milwaukee Co.,S-D2,D
Milwaukee Co.,S-D3,D
Milwaukee Co.,S-D4,D
Milwaukee Co.,S-D4,D

1 city holc_id HOLC Grade (manually mapped)
37 Milwaukee Co. S-D1 D
38 Milwaukee Co. S-D2 D
39 Milwaukee Co. S-D3 D
40 Milwaukee Co. S-D4 D

114
data/data-pipeline/data_pipeline/etl/sources/mapping_inequality/etl.py
View file

@ -1,10 +1,12 @@
import pathlib
import numpy as np
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.score import field_names
from data_pipeline.utils import download_file_from_url, get_module_logger
from data_pipeline.utils import download_file_from_url
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
logger = get_module_logger(__name__)
@ -21,10 +23,16 @@ class MappingInequalityETL(ExtractTransformLoad):
"""
def __init__(self):
self.MAPPING_INEQUALITY_CSV_URL = (
"https://raw.githubusercontent.com/americanpanorama/Census_HOLC_Research/"
"main/2010_Census_Tracts/holc_tract_lookup.csv"
)
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
self.MAPPING_INEQUALITY_CSV_URL = (
f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/"
"mapping_inequality/holc_tract_lookup.csv"
)
else:
self.MAPPING_INEQUALITY_CSV_URL = (
"https://raw.githubusercontent.com/americanpanorama/Census_HOLC_Research/"
"main/2010_Census_Tracts/holc_tract_lookup.csv"
)
self.MAPPING_INEQUALITY_CSV = (
self.get_tmp_path() / "holc_tract_lookup.csv"
)
@ -47,16 +55,21 @@ class MappingInequalityETL(ExtractTransformLoad):
self.HOLC_GRADE_AND_ID_FIELD: str = "holc_id"
self.CITY_INPUT_FIELD: str = "city"
self.HOLC_GRADE_D_FIELD: str = "HOLC Grade D"
self.HOLC_GRADE_D_FIELD: str = "HOLC Grade D (hazardous)"
self.HOLC_GRADE_C_FIELD: str = "HOLC Grade C (declining)"
self.HOLC_GRADE_MANUAL_FIELD: str = "HOLC Grade (manually mapped)"
self.HOLC_GRADE_DERIVED_FIELD: str = "HOLC Grade (derived)"
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
field_names.HOLC_GRADE_C_TRACT_PERCENT_FIELD,
field_names.HOLC_GRADE_C_OR_D_TRACT_PERCENT_FIELD,
field_names.HOLC_GRADE_C_OR_D_TRACT_50_PERCENT_FIELD,
field_names.HOLC_GRADE_D_TRACT_PERCENT_FIELD,
field_names.HOLC_GRADE_D_TRACT_20_PERCENT_FIELD,
field_names.HOLC_GRADE_D_TRACT_50_PERCENT_FIELD,
field_names.HOLC_GRADE_D_TRACT_75_PERCENT_FIELD,
field_names.REDLINED_SHARE,
]
self.df: pd.DataFrame
@ -113,34 +126,58 @@ class MappingInequalityETL(ExtractTransformLoad):
how="left",
)
# Create a single field that combines the 'derived' grade D field with the
# manually mapped grade D field into a single grade D field.
merged_df[self.HOLC_GRADE_D_FIELD] = np.where(
(merged_df[self.HOLC_GRADE_DERIVED_FIELD] == "D")
| (merged_df[self.HOLC_GRADE_MANUAL_FIELD] == "D"),
True,
None,
)
# Create a single field that combines the 'derived' grade C and D fields with the
# manually mapped grade C and D field into a single grade C and D field.
## Note: there are no manually derived C tracts at the moment
# Start grouping by, to sum all of the grade D parts of each tract.
grouped_df = (
merged_df.groupby(
by=[
self.GEOID_TRACT_FIELD_NAME,
self.HOLC_GRADE_D_FIELD,
],
# Keep the nulls, so we know the non-D proportion.
dropna=False,
)[self.TRACT_PROPORTION_FIELD]
for grade, field_name in [
("C", self.HOLC_GRADE_C_FIELD),
("D", self.HOLC_GRADE_D_FIELD),
]:
merged_df[field_name] = np.where(
(merged_df[self.HOLC_GRADE_DERIVED_FIELD] == grade)
| (merged_df[self.HOLC_GRADE_MANUAL_FIELD] == grade),
True,
None,
)
redlined_dataframes_list = [
merged_df[merged_df[field].fillna(False)]
.groupby(self.GEOID_TRACT_FIELD_NAME)[self.TRACT_PROPORTION_FIELD]
.sum()
.rename(new_name)
for field, new_name in [
(
self.HOLC_GRADE_D_FIELD,
field_names.HOLC_GRADE_D_TRACT_PERCENT_FIELD,
),
(
self.HOLC_GRADE_C_FIELD,
field_names.HOLC_GRADE_C_TRACT_PERCENT_FIELD,
),
]
]
# Group by tract ID to get tract proportions of just C or just D
# This produces a single row per tract
grouped_df = (
pd.concat(
redlined_dataframes_list,
axis=1,
)
.fillna(0)
.reset_index()
)
# Create a field that is only the percent that is grade D.
grouped_df[field_names.HOLC_GRADE_D_TRACT_PERCENT_FIELD] = np.where(
grouped_df[self.HOLC_GRADE_D_FIELD],
grouped_df[self.TRACT_PROPORTION_FIELD],
0,
grouped_df[
field_names.HOLC_GRADE_C_OR_D_TRACT_PERCENT_FIELD
] = grouped_df[
[
field_names.HOLC_GRADE_C_TRACT_PERCENT_FIELD,
field_names.HOLC_GRADE_D_TRACT_PERCENT_FIELD,
]
].sum(
axis=1
)
# Calculate some specific threshold cutoffs, for convenience.
@ -154,15 +191,14 @@ class MappingInequalityETL(ExtractTransformLoad):
grouped_df[field_names.HOLC_GRADE_D_TRACT_PERCENT_FIELD] > 0.75
)
# Drop the non-True values of `self.HOLC_GRADE_D_FIELD` -- we only
# want one row per tract for future joins.
# Note this means not all tracts will be in this data.
# Note: this singleton comparison warning may be a pylint bug:
# https://stackoverflow.com/questions/51657715/pylint-pandas-comparison-to-true-should-be-just-expr-or-expr-is-true-sin#comment90876517_51657715
# pylint: disable=singleton-comparison
grouped_df = grouped_df[
grouped_df[self.HOLC_GRADE_D_FIELD] == True # noqa: E712
]
grouped_df[field_names.HOLC_GRADE_C_OR_D_TRACT_50_PERCENT_FIELD] = (
grouped_df[field_names.HOLC_GRADE_C_OR_D_TRACT_PERCENT_FIELD] > 0.5
)
# Create the indicator we will use
grouped_df[field_names.REDLINED_SHARE] = (
grouped_df[field_names.HOLC_GRADE_C_OR_D_TRACT_PERCENT_FIELD] > 0.5
) & (grouped_df[field_names.HOLC_GRADE_D_TRACT_PERCENT_FIELD] > 0)
# Sort for convenience.
grouped_df.sort_values(by=self.GEOID_TRACT_FIELD_NAME, inplace=True)

4
data/data-pipeline/data_pipeline/etl/sources/maryland_ejscreen/README.md
View file

@ -8,7 +8,7 @@ According to the documentation:
There exist two data categories: Population Burden and Population Characteristics.
There are two indicators within Population Burden: Exposure, and Socioeconomic. Within Population Characteristics, there exist two indicators: Sensitive, Environmental Effects. Each respective indicator contains several relevant covariates, and an averaged score.
There are two indicators within Population Burden: Exposure, and Socioeconomic. Within Population Characteristics, there exist two indicators: Sensitive, Environmental Effects. Each respective indicator contains several relevant covariates, and an averaged score.
The two "Pollution Burden" average scores are then averaged together and the result is multiplied by the average of the "Population Characteristics" categories to get the total EJ Score for each tract.
@ -20,4 +20,4 @@ Furthermore, it was determined that Bladensburg residents are at a higher risk o
Source:
Driver, A.; Mehdizadeh, C.; Bara-Garcia, S.; Bodenreider, C.; Lewis, J.; Wilson, S. Utilization of the Maryland Environmental Justice Screening Tool: A Bladensburg, Maryland Case Study. Int. J. Environ. Res. Public Health 2019, 16, 348.
Driver, A.; Mehdizadeh, C.; Bara-Garcia, S.; Bodenreider, C.; Lewis, J.; Wilson, S. Utilization of the Maryland Environmental Justice Screening Tool: A Bladensburg, Maryland Case Study. Int. J. Environ. Res. Public Health 2019, 16, 348.

8
data/data-pipeline/data_pipeline/etl/sources/maryland_ejscreen/etl.py
View file

@ -1,11 +1,11 @@
from glob import glob
import geopandas as gpd
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger
from data_pipeline.score import field_names
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)

3
data/data-pipeline/data_pipeline/etl/sources/michigan_ejscreen/README.md
View file

@ -1,5 +1,4 @@
# Michigan EJSCREEN
<!-- markdown-link-check-disable -->
The Michigan EJSCREEN description and publication can be found [here](https://deepblue.lib.umich.edu/bitstream/handle/2027.42/149105/AssessingtheStateofEnvironmentalJusticeinMichigan_344.pdf).
<!-- markdown-link-check-enable-->
@ -30,4 +29,4 @@ Sources:
* Minnesota Pollution Control Agency. (2015, December 15). Environmental Justice Framework Report.
Retrieved from https://www.pca.state.mn.us/sites/default/files/p-gen5-05.pdf.
* Faust, J., L. August, K. Bangia, V. Galaviz, J. Leichty, S. Prasad… and L. Zeise. (2017, January). Update to the California Communities Environmental Health Screening Tool CalEnviroScreen 3.0. Retrieved from OEHHA website: https://oehha.ca.gov/media/downloads/calenviroscreen/report/ces3report.pdf
* Faust, J., L. August, K. Bangia, V. Galaviz, J. Leichty, S. Prasad… and L. Zeise. (2017, January). Update to the California Communities Environmental Health Screening Tool CalEnviroScreen 3.0. Retrieved from OEHHA website: https://oehha.ca.gov/media/downloads/calenviroscreen/report/ces3report.pdf

7
data/data-pipeline/data_pipeline/etl/sources/michigan_ejscreen/etl.py
View file

@ -1,9 +1,8 @@
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger
from data_pipeline.score import field_names
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)

45
data/data-pipeline/data_pipeline/etl/sources/national_risk_index/etl.py
View file

@ -2,11 +2,11 @@
# but it may be a known bug. https://github.com/PyCQA/pylint/issues/1498
# pylint: disable=unsubscriptable-object
# pylint: disable=unsupported-assignment-operation
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad, ValidGeoLevel
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
logger = get_module_logger(__name__)
@ -15,8 +15,21 @@ class NationalRiskIndexETL(ExtractTransformLoad):
"""ETL class for the FEMA National Risk Index dataset"""
NAME = "national_risk_index"
SOURCE_URL = "https://hazards.fema.gov/nri/Content/StaticDocuments/DataDownload//NRI_Table_CensusTracts/NRI_Table_CensusTracts.zip"
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
SOURCE_URL = (
f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/"
"national_risk_index/NRI_Table_CensusTracts.zip"
)
else:
SOURCE_URL = (
"https://hazards.fema.gov/nri/Content/StaticDocuments/DataDownload/"
"NRI_Table_CensusTracts/NRI_Table_CensusTracts.zip"
)
GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
PUERTO_RICO_EXPECTED_IN_DATA = False
LOAD_YAML_CONFIG: bool = True
# Output score variables (values set on datasets.yml) for linting purposes
RISK_INDEX_EXPECTED_ANNUAL_LOSS_SCORE_FIELD_NAME: str
@ -33,9 +46,6 @@ class NationalRiskIndexETL(ExtractTransformLoad):
AGRIVALUE_LOWER_BOUND = 408000
def __init__(self):
# load YAML config
self.DATASET_CONFIG = super().yaml_config_load()
# define the full path for the input CSV file
self.INPUT_CSV = self.get_tmp_path() / "NRI_Table_CensusTracts.csv"
@ -156,6 +166,27 @@ class NationalRiskIndexETL(ExtractTransformLoad):
lower=self.AGRIVALUE_LOWER_BOUND
)
## Check that this clip worked -- that the only place the value has changed is when the clip took effect
base_expectation = (
disaster_agriculture_sum_series
/ df_nri[self.AGRICULTURAL_VALUE_INPUT_FIELD_NAME]
)
assert (
df_nri[
df_nri[self.EXPECTED_AGRICULTURE_LOSS_RATE_FIELD_NAME]
!= base_expectation
][self.AGRICULTURAL_VALUE_INPUT_FIELD_NAME].max()
<= self.AGRIVALUE_LOWER_BOUND
), (
"Clipping the agrivalue did not work. There are places where the value doesn't "
+ "match an unclipped ratio, even where the agrivalue is above the lower bound!"
)
assert (
df_nri[self.EXPECTED_AGRICULTURE_LOSS_RATE_FIELD_NAME]
!= base_expectation
).sum() > 0, "Clipping the agrivalue did nothing!"
# This produces a boolean that is True in the case of non-zero agricultural value
df_nri[self.CONTAINS_AGRIVALUE] = (
df_nri[self.AGRICULTURAL_VALUE_INPUT_FIELD_NAME] > 0

80
data/data-pipeline/data_pipeline/etl/sources/nlcd_nature_deprived/README.md Normal file
View file

@ -0,0 +1,80 @@
# Nature deprived communities data
The following dataset was compiled by TPL (Trust for Public Lands) using NCLD data. We define as: AREA - [CROPLAND] - [IMPERVIOUS SURFACES].
## Codebook
- GEOID10 – Census tract ID
- SF – State Name
- CF – County Name
- P200_PFS – Percent of individuals below 200% Federal Poverty Line (from CEJST source data).
- CA_LT20 – Percent higher ed enrollment rate is less than 20% (from CEJST source data).
- TractAcres – Acres of tract calculated from ALAND10 field (area land/meters) in 2010 census tracts.
- CAVEAT: Some census tracts in the CEJST source file extend into open water. ALAND10 area was used to constrain percent calculations (e.g. cropland area) to land only.
- AcresCrops – Acres crops calculated by summing all cells in the NLCD Cropland Data Layer crop classes.
- PctCrops – Formula: AcresCrops/TractAcres*100.
- PctImperv – Mean imperviousness for each census tract.
- CAVEAT: Where tracts extend into open water, mean imperviousness may be underestimated.
- __TO USE__ PctNatural – Formula: 100 – PctCrops – PctImperv.
- PctNat90 – Tract in or below 10th percentile for PctNatural. 1 = True, 0 = False.
- PctNatural 10th percentile = 28.6439%
- ImpOrCrop – If tract >= 90th percentile for PctImperv OR PctCrops. 1 = True, 0 = False.
- PctImperv 90th percentile = 67.4146 %
- PctCrops 90th percentile = 27.8116 %
- LowInAndEd – If tract >= 65th percentile for P200_PFS AND CA_LT20.
- P200_PFS 65th percentile = 64.0%
- NatureDep – ImpOrCrp = 1 AND LowInAndEd = 1.
We added `GEOID10_TRACT` before converting shapefile to csv.
## Instructions to recreate
### Creating Impervious plus Cropland Attributes for Census Tracts
The Cropland Data Layer and NLCD Impervious layer were too big to put on our OneDrive, but you can download them here:
CDL: https://www.nass.usda.gov/Research_and_Science/Cropland/Release/datasets/2021_30m_cdls.zip
Impervious: https://s3-us-west-2.amazonaws.com/mrlc/nlcd_2019_impervious_l48_20210604.zip
#### Crops
Add an attribute called TractAcres (or similar) to the census tracts to hold a value representing acres covered by the census tract.
Calculate the TractAcres field for each census tract by using the Calculate Geometry tool (set the Property to Area (geodesic), and the Units to Acres).
From the Cropland Data Layer (CDL), extract only the pixels representing crops, using the Extract by Attributes tool in ArcGIS Spatial Analyst toolbox.
a. The attribute table tells you the names of each type of land cover. Since the CDL also contains NLCD classes and empty classes, the actual crop classes must be extracted.
From the crops-only raster extracted from the CDL, run the Reclassify tool to create a binary layer where all crops = 1, and everything else is Null.
Run the Tabulate Area tool:
a. Zone data = census tracts
b. Input raster data = the binary crops layer
c. This will produce a table with the square meters of crops in each census tract contained in an attribute called VALUE_1
Run the Join Field tool to join the table to the census tracts, with the VALUE_1 field as the Transfer Field, to transfer the VALUE_1 field (square meters of crops) to the census tracts.
Add a field to the census tracts called AcresCrops (or similar) to hold the acreage of crops in each census tract.
Calculate the AcresCrops field by multiplying the VALUE_1 field by 0.000247105 to produce acres of crops in each census tracts.
a. You can delete the VALUE_1 field.
Add a field called PctCrops (or similar) to hold the percent of each census tract occupied by crops.
Calculate the PctCrops field by dividing the AcresCrops field by the TractAcres field, and multiply by 100 to get the percent.
Impervious
Run the Zonal Statistics as Table tool:
a. Zone data = census tracts
b. Input raster data = impervious data raster layer
c. Statistics type = Mean
d. This will produce a table with the percent of each census tract occupied by impervious surfaces, contained in an attribute called MEAN
Run the Join Field tool to join the table to the census tracts, with the MEAN field as the Transfer Field, to transfer the MEAN field (percent impervious) to the census tracts.
Add a field called PctImperv (or similar) to hold the percent impervious value.
Calculate the PctImperv field by setting it equal to the MEAN field.
a. You can delete the MEAN field.
Combine the Crops and Impervious Data
Open the census tracts attribute table and add a field called PctNatural (or similar). Calculate this field using this equation: 100 – PctCrops – PctImperv . This produces a value that tells you the percent of each census tract covered in natural land cover.
Define the census tracts that fall in the 90th percentile of non-natural land cover:
a. Add a field called PctNat90 (or similar)
b. Right-click on the PctNatural field, and click Sort Ascending (lowest PctNatural values on top)
c. Select the top 10 percent of rows after the sort
d. Click on Show Selected Records in the attribute table
e. Calculate the PctNat90 field for the selected records = 1
f. Clear the selection
g. The rows that now have a value of 1 for PctNat90 are the most lacking for natural land cover, and can be symbolized accordingly in a map

0
data/data-pipeline/data_pipeline/etl/sources/nlcd_nature_deprived/__init__.py Normal file
View file

77
data/data-pipeline/data_pipeline/etl/sources/nlcd_nature_deprived/etl.py Normal file
View file

@ -0,0 +1,77 @@
# pylint: disable=unsubscriptable-object
# pylint: disable=unsupported-assignment-operation
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)
class NatureDeprivedETL(ExtractTransformLoad):
"""ETL class for the Nature Deprived Communities dataset"""
NAME = "nlcd_nature_deprived"
SOURCE_URL = (
settings.AWS_JUSTICE40_DATASOURCES_URL
+ "/usa_conus_nat_dep__compiled_by_TPL.csv.zip"
)
GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
PUERTO_RICO_EXPECTED_IN_DATA = False
LOAD_YAML_CONFIG: bool = True
ALASKA_AND_HAWAII_EXPECTED_IN_DATA = False
# Output score variables (values set on datasets.yml) for linting purposes
ELIGIBLE_FOR_NATURE_DEPRIVED_FIELD_NAME: str
TRACT_PERCENT_IMPERVIOUS_FIELD_NAME: str
TRACT_PERCENT_NON_NATURAL_FIELD_NAME: str
TRACT_PERCENT_CROPLAND_FIELD_NAME: str
def __init__(self):
# define the full path for the input CSV file
self.INPUT_CSV = (
self.get_tmp_path() / "usa_conus_nat_dep__compiled_by_TPL.csv"
)
# this is the main dataframe
self.df: pd.DataFrame
# Start dataset-specific vars here
self.PERCENT_NATURAL_FIELD_NAME = "PctNatural"
self.PERCENT_IMPERVIOUS_FIELD_NAME = "PctImperv"
self.PERCENT_CROPLAND_FIELD_NAME = "PctCrops"
self.TRACT_ACRES_FIELD_NAME = "TractAcres"
# In order to ensure that tracts with very small Acreage, we want to create an eligibility criterion
# similar to agrivalue. Here, we are ensuring that a tract has at least 35 acres, or is above the 1st percentile
# for area. This does indeed remove tracts from the 90th+ percentile later on
self.TRACT_ACRES_LOWER_BOUND = 35
def transform(self) -> None:
"""Reads the unzipped data file into memory and applies the following
transformations to prepare it for the load() method:
- Renames columns as needed
"""
logger.info("Transforming NLCD Data")
df_ncld: pd.DataFrame = pd.read_csv(
self.INPUT_CSV,
dtype={self.INPUT_GEOID_TRACT_FIELD_NAME: str},
low_memory=False,
)
df_ncld[self.ELIGIBLE_FOR_NATURE_DEPRIVED_FIELD_NAME] = (
df_ncld[self.TRACT_ACRES_FIELD_NAME] >= self.TRACT_ACRES_LOWER_BOUND
)
df_ncld[self.TRACT_PERCENT_NON_NATURAL_FIELD_NAME] = (
100 - df_ncld[self.PERCENT_NATURAL_FIELD_NAME]
)
# Assign the final df to the class' output_df for the load method with rename
self.output_df = df_ncld.rename(
columns={
self.PERCENT_IMPERVIOUS_FIELD_NAME: self.TRACT_PERCENT_IMPERVIOUS_FIELD_NAME,
self.PERCENT_CROPLAND_FIELD_NAME: self.TRACT_PERCENT_CROPLAND_FIELD_NAME,
}
)

27
data/data-pipeline/data_pipeline/etl/sources/persistent_poverty/etl.py
View file

@ -1,12 +1,11 @@
import functools
import pandas as pd
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import (
get_module_logger,
unzip_file_from_url,
)
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import unzip_file_from_url
logger = get_module_logger(__name__)
@ -19,6 +18,10 @@ class PersistentPovertyETL(ExtractTransformLoad):
Codebook: `https://s4.ad.brown.edu/Projects/Diversity/Researcher/LTBDDload/Dfiles/codebooks.pdf`.
"""
NAME = "persistent_poverty"
GEO_LEVEL: ValidGeoLevel = ValidGeoLevel.CENSUS_TRACT
PUERTO_RICO_EXPECTED_IN_DATA = False
def __init__(self):
self.OUTPUT_PATH = self.DATA_PATH / "dataset" / "persistent_poverty"
@ -75,7 +78,7 @@ class PersistentPovertyETL(ExtractTransformLoad):
def extract(self) -> None:
logger.info("Starting to download 86MB persistent poverty file.")
unzipped_file_path = self.get_tmp_path() / "persistent_poverty"
unzipped_file_path = self.get_tmp_path()
unzip_file_from_url(
file_url=settings.AWS_JUSTICE40_DATASOURCES_URL
@ -155,14 +158,4 @@ class PersistentPovertyETL(ExtractTransformLoad):
)
)
self.df = transformed_df
def load(self) -> None:
logger.info("Saving persistent poverty data.")
# mkdir census
self.OUTPUT_PATH.mkdir(parents=True, exist_ok=True)
self.df[self.COLUMNS_TO_KEEP].to_csv(
path_or_buf=self.OUTPUT_PATH / "usa.csv", index=False
)
self.output_df = transformed_df

99
data/data-pipeline/data_pipeline/etl/sources/tribal/etl.py
View file

@ -1,18 +1,25 @@
from pathlib import Path
import geopandas as gpd
import pandas as pd
from data_pipeline.config import settings
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.utils import get_module_logger, unzip_file_from_url
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import unzip_file_from_url
logger = get_module_logger(__name__)
class TribalETL(ExtractTransformLoad):
def __init__(self):
self.GEOJSON_BASE_PATH = self.DATA_PATH / "tribal" / "geojson"
self.GEOGRAPHIC_BASE_PATH = (
self.DATA_PATH / "tribal" / "geographic_data"
)
self.CSV_BASE_PATH = self.DATA_PATH / "tribal" / "csv"
self.NATIONAL_TRIBAL_GEOJSON_PATH = self.GEOJSON_BASE_PATH / "usa.json"
self.NATIONAL_TRIBAL_GEOJSON_PATH = (
self.GEOGRAPHIC_BASE_PATH / "usa.json"
)
self.USA_TRIBAL_DF_LIST = []
def extract(self) -> None:
@ -23,43 +30,66 @@ class TribalETL(ExtractTransformLoad):
"""
logger.info("Downloading Tribal Data")
bia_geojson_url = "https://justice40-data.s3.amazonaws.com/data-sources/BIA_National_LAR_json.zip"
alaska_geojson_url = "https://justice40-data.s3.amazonaws.com/data-sources/Alaska_Native_Villages_json.zip"
bia_shapefile_zip_url = (
settings.AWS_JUSTICE40_DATASOURCES_URL
+ "/BIA_National_LAR_updated_20220929.zip"
)
tsa_and_aian_geojson_zip_url = (
settings.AWS_JUSTICE40_DATASOURCES_URL
+ "/BIA_TSA_and_AIAN_json.zip"
)
alaska_geojson_url = (
settings.AWS_JUSTICE40_DATASOURCES_URL
+ "/Alaska_Native_Villages_json.zip"
)
unzip_file_from_url(
bia_geojson_url,
bia_shapefile_zip_url,
self.TMP_PATH,
self.DATA_PATH / "tribal" / "geojson" / "bia_national_lar",
self.GEOGRAPHIC_BASE_PATH / "bia_national_lar",
)
unzip_file_from_url(
tsa_and_aian_geojson_zip_url,
self.TMP_PATH,
self.GEOGRAPHIC_BASE_PATH / "tsa_and_aian",
)
unzip_file_from_url(
alaska_geojson_url,
self.TMP_PATH,
self.DATA_PATH / "tribal" / "geojson" / "alaska_native_villages",
self.GEOGRAPHIC_BASE_PATH / "alaska_native_villages",
)
pass
def _transform_bia_national_lar(self, tribal_geojson_path: Path) -> None:
def _transform_bia_national_lar(self, path: Path) -> None:
"""Transform the Tribal BIA National Lar Geodataframe and appends it to the
national Tribal Dataframe List
Args:
tribal_geojson_path (Path): the Path to the Tribal Geojson
path (Path): the Path to the BIA National Lar
Returns:
None
"""
bia_national_lar_df = gpd.read_file(tribal_geojson_path)
bia_national_lar_df = gpd.read_file(path)
# DELETE
logger.info(f"Columns: {bia_national_lar_df.columns}\n")
bia_national_lar_df.drop(
["OBJECTID", "GISAcres", "Shape_Length", "Shape_Area"],
["GISAcres"],
axis=1,
inplace=True,
)
bia_national_lar_df.rename(
columns={"TSAID": "tribalId", "LARName": "landAreaName"},
columns={
"LARID": field_names.TRIBAL_ID,
"LARName": field_names.TRIBAL_LAND_AREA_NAME,
},
inplace=True,
)
@ -87,7 +117,10 @@ class TribalETL(ExtractTransformLoad):
)
bia_aian_supplemental_df.rename(
columns={"OBJECTID": "tribalId", "Land_Area_": "landAreaName"},
columns={
"OBJECTID": field_names.TRIBAL_ID,
"Land_Area_": field_names.TRIBAL_LAND_AREA_NAME,
},
inplace=True,
)
@ -113,7 +146,10 @@ class TribalETL(ExtractTransformLoad):
)
bia_tsa_df.rename(
columns={"TSAID": "tribalId", "LARName": "landAreaName"},
columns={
"TSAID": field_names.TRIBAL_ID,
"LARName": field_names.TRIBAL_LAND_AREA_NAME,
},
inplace=True,
)
@ -136,8 +172,8 @@ class TribalETL(ExtractTransformLoad):
alaska_native_villages_df.rename(
columns={
"GlobalID": "tribalId",
"TRIBALOFFICENAME": "landAreaName",
"GlobalID": field_names.TRIBAL_ID,
"TRIBALOFFICENAME": field_names.TRIBAL_LAND_AREA_NAME,
},
inplace=True,
)
@ -152,27 +188,30 @@ class TribalETL(ExtractTransformLoad):
"""
logger.info("Transforming Tribal Data")
# load the geojsons
bia_national_lar_geojson = (
self.GEOJSON_BASE_PATH / "bia_national_lar" / "BIA_TSA.json"
# Set the filepaths:
bia_national_lar_shapefile = (
self.GEOGRAPHIC_BASE_PATH / "bia_national_lar"
)
bia_aian_supplemental_geojson = (
self.GEOJSON_BASE_PATH
/ "bia_national_lar"
self.GEOGRAPHIC_BASE_PATH
/ "tsa_and_aian"
/ "BIA_AIAN_Supplemental.json"
)
bia_tsa_geojson_geojson = (
self.GEOJSON_BASE_PATH / "bia_national_lar" / "BIA_TSA.json"
bia_tsa_geojson = (
self.GEOGRAPHIC_BASE_PATH / "tsa_and_aian" / "BIA_TSA.json"
)
alaska_native_villages_geojson = (
self.GEOJSON_BASE_PATH
self.GEOGRAPHIC_BASE_PATH
/ "alaska_native_villages"
/ "AlaskaNativeVillages.gdb.geojson"
)
self._transform_bia_national_lar(bia_national_lar_geojson)
self._transform_bia_national_lar(bia_national_lar_shapefile)
self._transform_bia_aian_supplemental(bia_aian_supplemental_geojson)
self._transform_bia_tsa(bia_tsa_geojson_geojson)
self._transform_bia_tsa(bia_tsa_geojson)
self._transform_alaska_native_villages(alaska_native_villages_geojson)
def load(self) -> None:
@ -182,13 +221,13 @@ class TribalETL(ExtractTransformLoad):
None
"""
logger.info("Saving Tribal GeoJson and CSV")
usa_tribal_df = gpd.GeoDataFrame(
pd.concat(self.USA_TRIBAL_DF_LIST, ignore_index=True)
)
usa_tribal_df = usa_tribal_df.to_crs(
"+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs"
)
logger.info("Writing national geojson file")
usa_tribal_df.to_file(
self.NATIONAL_TRIBAL_GEOJSON_PATH, driver="GeoJSON"

9
data/data-pipeline/data_pipeline/etl/sources/tribal/etl_utils.py
View file

@ -1,11 +1,8 @@
from pathlib import Path
from data_pipeline.utils import (
get_module_logger,
remove_all_from_dir,
remove_files_from_dir,
)
from data_pipeline.utils import get_module_logger
from data_pipeline.utils import remove_all_from_dir
from data_pipeline.utils import remove_files_from_dir
logger = get_module_logger(__name__)

0
data/data-pipeline/data_pipeline/etl/sources/tribal_overlap/README.md Normal file
View file

0
data/data-pipeline/data_pipeline/etl/sources/tribal_overlap/__init__.py Normal file
View file

274
data/data-pipeline/data_pipeline/etl/sources/tribal_overlap/etl.py Normal file
View file

@ -0,0 +1,274 @@
from typing import Optional
import geopandas as gpd
import numpy as np
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.etl.sources.geo_utils import add_tracts_for_geometries
from data_pipeline.etl.sources.geo_utils import get_tract_geojson
from data_pipeline.etl.sources.geo_utils import get_tribal_geojson
from data_pipeline.score import field_names
from data_pipeline.utils import get_module_logger
logger = get_module_logger(__name__)
class TribalOverlapETL(ExtractTransformLoad):
"""Calculates the overlap between Census tracts and Tribal boundaries."""
# Metadata for the baseclass
NAME = "tribal_overlap"
GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
PUERTO_RICO_EXPECTED_IN_DATA = False
ALASKA_AND_HAWAII_EXPECTED_IN_DATA = True
EXPECTED_MISSING_STATES = [
# 15 is Hawaii, which has Hawaiian Home Lands, but they are not included in
# this dataset.
"15",
# The following states do not have any federally recognized Tribes in this
# dataset.
"10",
"11",
"13",
"17",
"18",
"21",
"24",
"33",
"34",
"39",
"50",
"51",
"54",
]
# A Tribal area that requires some special processing.
ANNETTE_ISLAND_TRIBAL_NAME = "Annette Island"
CRS_INTEGER = 3857
TRIBAL_OVERLAP_CUTOFF = 0.995 # Percentage of overlap that rounds to 100%
# Define these for easy code completion
def __init__(self):
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
field_names.COUNT_OF_TRIBAL_AREAS_IN_TRACT_AK,
field_names.COUNT_OF_TRIBAL_AREAS_IN_TRACT_CONUS,
field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT,
field_names.NAMES_OF_TRIBAL_AREAS_IN_TRACT,
field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT_DISPLAY,
field_names.IS_TRIBAL_DAC,
]
self.OVERALL_TRIBAL_COUNT = "OVERALL_TRIBAL_COUNT"
self.output_df: pd.DataFrame
self.census_tract_gdf: gpd.GeoDataFrame
self.tribal_gdf: gpd.GeoDataFrame
@staticmethod
def _create_string_from_list(series: pd.Series) -> str:
"""Helper method that creates a sorted string list (for tribal names)."""
str_list = series.tolist()
str_list = sorted(str_list)
return ", ".join(str_list)
@classmethod
def _adjust_percentage_for_frontend(
cls,
percentage_float: float,
) -> Optional[float]:
"""Round numbers very close to 0 to 0 and very close to 1 to 1 for display"""
if percentage_float is None:
return None
if percentage_float < 0.01:
return 0.0
if percentage_float > cls.TRIBAL_OVERLAP_CUTOFF:
return 1.0
return percentage_float
def extract(self) -> None:
self.census_tract_gdf = get_tract_geojson()
self.tribal_gdf = get_tribal_geojson()
def transform(self) -> None:
logger.info("Starting tribal overlap transforms.")
# First, calculate whether tracts include any areas from the Tribal areas,
# for both the points in AK and the polygons in the continental US (CONUS).
tribal_overlap_with_tracts = add_tracts_for_geometries(
df=self.tribal_gdf, tract_data=self.census_tract_gdf
)
# Cleanup the suffixes in the tribal names
tribal_overlap_with_tracts[field_names.TRIBAL_LAND_AREA_NAME] = (
tribal_overlap_with_tracts[field_names.TRIBAL_LAND_AREA_NAME]
.str.replace(" LAR", "")
.str.replace(" TSA", "")
.str.replace(" IRA", "")
.str.replace(" AK", "")
)
tribal_overlap_with_tracts = tribal_overlap_with_tracts.groupby(
[self.GEOID_TRACT_FIELD_NAME]
).agg(
{
field_names.TRIBAL_ID: "count",
field_names.TRIBAL_LAND_AREA_NAME: self._create_string_from_list,
}
)
tribal_overlap_with_tracts = tribal_overlap_with_tracts.reset_index()
tribal_overlap_with_tracts = tribal_overlap_with_tracts.rename(
columns={
field_names.TRIBAL_ID: self.OVERALL_TRIBAL_COUNT,
field_names.TRIBAL_LAND_AREA_NAME: field_names.NAMES_OF_TRIBAL_AREAS_IN_TRACT,
}
)
# Second, calculate percentage overlap.
# Drop the points from the Tribal data (because these cannot be joined to a
# (Multi)Polygon tract data frame)
tribal_gdf_without_points = self.tribal_gdf[
self.tribal_gdf.geom_type.isin(["Polygon", "MultiPolygon"])
]
# Switch from geographic to projected CRSes
# because logically that's right
self.census_tract_gdf = self.census_tract_gdf.to_crs(
crs=self.CRS_INTEGER
)
tribal_gdf_without_points = tribal_gdf_without_points.to_crs(
crs=self.CRS_INTEGER
)
# Create a measure for the entire census tract area
self.census_tract_gdf["area_tract"] = self.census_tract_gdf.area
# Performing overlay funcion
# We have a mix of polygons and multipolygons, and we just want the overlaps
# without caring a ton about the specific types, so we ignore geom type.
# Realistically, this changes almost nothing in the calculation; True and False
# are the same within 9 digits of precision
gdf_joined = gpd.overlay(
self.census_tract_gdf,
tribal_gdf_without_points,
how="intersection",
keep_geom_type=False,
)
# Calculating the areas of the newly-created overlapping geometries
gdf_joined["area_joined"] = gdf_joined.area
# Calculating the areas of the newly-created geometries in relation
# to the original tract geometries
gdf_joined[field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT] = (
gdf_joined["area_joined"] / gdf_joined["area_tract"]
)
# Aggregate the results
percentage_results = gdf_joined.groupby(
[self.GEOID_TRACT_FIELD_NAME]
).agg({field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT: "sum"})
percentage_results = percentage_results.reset_index()
# Merge the two results.
merged_output_df = tribal_overlap_with_tracts.merge(
right=percentage_results,
how="outer",
on=self.GEOID_TRACT_FIELD_NAME,
)
# Finally, fix one unique error.
# There is one unique Tribal area (self.ANNETTE_ISLAND_TRIBAL_NAME) that is a polygon in
# Alaska. All other Tribal areas in Alaska are points.
# For tracts that *only* contain that Tribal area, leave percentage as is.
# For tracts that include that Tribal area AND Alaska Native villages,
# null the percentage, because we cannot calculate the percent of the tract
# this is within Tribal areas.
# Create state FIPS codes.
merged_output_df_state_fips_code = merged_output_df[
self.GEOID_TRACT_FIELD_NAME
].str[0:2]
# Start by testing for Annette Island exception, to make sure data is as
# expected
alaskan_non_annette_matches = (
# Data from Alaska
(merged_output_df_state_fips_code == "02")
# Where the Tribal areas do *not* include Annette
& (
~merged_output_df[
field_names.NAMES_OF_TRIBAL_AREAS_IN_TRACT
].str.contains(self.ANNETTE_ISLAND_TRIBAL_NAME)
)
# But somehow percentage is greater than zero.
& (
merged_output_df[field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT]
> 0
)
)
# There should be none of these matches.
if sum(alaskan_non_annette_matches) > 0:
raise ValueError(
"Data has changed. More than one Alaskan Tribal Area has polygon "
"boundaries. You'll need to refactor this ETL. \n"
f"Data:\n{merged_output_df[alaskan_non_annette_matches]}"
)
# Now, fix the exception that is already known.
merged_output_df[
field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT
] = np.where(
# For tracts inside Alaska
(merged_output_df_state_fips_code == "02")
# That are not only represented by Annette Island
& (
merged_output_df[field_names.NAMES_OF_TRIBAL_AREAS_IN_TRACT]
!= self.ANNETTE_ISLAND_TRIBAL_NAME
),
# Set the value to `None` for tracts with more than just Annette.
None,
# Otherwise, set the value to what it was.
merged_output_df[field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT],
)
# Counting tribes in the lower 48 is different from counting in AK,
# so per request by the design and frontend team, we remove all the
# counts outside AK
merged_output_df[
field_names.COUNT_OF_TRIBAL_AREAS_IN_TRACT_AK
] = np.where(
# In Alaska
(merged_output_df_state_fips_code == "02"),
# Keep the counts
merged_output_df[self.OVERALL_TRIBAL_COUNT],
# Otherwise, null them
None,
)
# TODO: Count tribal areas in the lower 48 correctly
merged_output_df[
field_names.COUNT_OF_TRIBAL_AREAS_IN_TRACT_CONUS
] = None
merged_output_df[field_names.IS_TRIBAL_DAC] = (
merged_output_df[field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT]
> self.TRIBAL_OVERLAP_CUTOFF
)
# The very final thing we want to do is produce a string for the front end to show
# We do this here so that all of the logic is included
merged_output_df[
field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT_DISPLAY
] = merged_output_df[field_names.PERCENT_OF_TRIBAL_AREA_IN_TRACT].apply(
self._adjust_percentage_for_frontend
)
self.output_df = merged_output_df

0
data/data-pipeline/data_pipeline/etl/sources/us_army_fuds/__init__.py Normal file
View file

112
data/data-pipeline/data_pipeline/etl/sources/us_army_fuds/etl.py Normal file
View file

@ -0,0 +1,112 @@
from pathlib import Path
import geopandas as gpd
import numpy as np
import pandas as pd
from data_pipeline.etl.base import ExtractTransformLoad
from data_pipeline.etl.base import ValidGeoLevel
from data_pipeline.etl.sources.geo_utils import add_tracts_for_geometries
from data_pipeline.utils import download_file_from_url
from data_pipeline.utils import get_module_logger
from data_pipeline.config import settings
logger = get_module_logger(__name__)
class USArmyFUDS(ExtractTransformLoad):
"""The Formerly Used Defense Sites (FUDS)"""
NAME: str = "us_army_fuds"
ELIGIBLE_FUDS_COUNT_FIELD_NAME: str
INELIGIBLE_FUDS_COUNT_FIELD_NAME: str
ELIGIBLE_FUDS_BINARY_FIELD_NAME: str
GEO_LEVEL: ValidGeoLevel = ValidGeoLevel.CENSUS_TRACT
LOAD_YAML_CONFIG: bool = True
ISLAND_AREAS_EXPECTED_IN_DATA = True
def __init__(self):
if settings.DATASOURCE_RETRIEVAL_FROM_AWS:
self.FILE_URL = (
f"{settings.AWS_JUSTICE40_DATASOURCES_URL}/raw-data-sources/"
"us_army_fuds/Formerly_Used_Defense_Sites_(FUDS)_"
"all_data_reported_to_Congress_in_FY2020.geojson"
)
else:
self.FILE_URL: str = (
"https://opendata.arcgis.com/api/v3/datasets/"
"3f8354667d5b4b1b8ad7a6e00c3cf3b1_1/downloads/"
"data?format=geojson&spatialRefId=4326&where=1%3D1"
)
self.OUTPUT_PATH: Path = self.DATA_PATH / "dataset" / "us_army_fuds"
# Constants for output
self.COLUMNS_TO_KEEP = [
self.GEOID_TRACT_FIELD_NAME,
self.ELIGIBLE_FUDS_COUNT_FIELD_NAME,
self.INELIGIBLE_FUDS_COUNT_FIELD_NAME,
self.ELIGIBLE_FUDS_BINARY_FIELD_NAME,
]
self.DOWNLOAD_FILE_NAME = self.get_tmp_path() / "fuds.geojson"
self.raw_df: gpd.GeoDataFrame
self.output_df: pd.DataFrame
def extract(self) -> None:
logger.info("Starting FUDS data download.")
download_file_from_url(
file_url=self.FILE_URL,
download_file_name=self.DOWNLOAD_FILE_NAME,
verify=True,
)
def transform(self) -> None:
logger.info("Starting FUDS transform.")
# before we try to do any transformation, get the tract data
# so it's loaded and the census ETL is out of scope
logger.info("Loading FUDS data as GeoDataFrame for transform")
raw_df = gpd.read_file(
filename=self.DOWNLOAD_FILE_NAME,
low_memory=False,
)
# Note that the length of raw_df will not be exactly the same
# because same bases lack coordinated or have coordinates in
# Mexico or in the ocean. See the following dataframe:
# raw_df[~raw_df.OBJECTID.isin(df_with_tracts.OBJECTID)][
# ['OBJECTID', 'CLOSESTCITY', 'COUNTY', 'ELIGIBILITY',
# 'STATE', 'LATITUDE', "LONGITUDE"]]
logger.debug("Adding tracts to FUDS data")
df_with_tracts = add_tracts_for_geometries(raw_df)
self.output_df = pd.DataFrame()
# this will create a boolean series which you can do actually sans np.where
df_with_tracts["tmp_fuds"] = (
df_with_tracts.ELIGIBILITY == "Eligible"
) & (df_with_tracts.HASPROJECTS == "Yes")
self.output_df[
self.ELIGIBLE_FUDS_COUNT_FIELD_NAME
] = df_with_tracts.groupby(self.GEOID_TRACT_FIELD_NAME)[
"tmp_fuds"
].sum()
self.output_df[self.INELIGIBLE_FUDS_COUNT_FIELD_NAME] = (
df_with_tracts[~df_with_tracts.tmp_fuds]
.groupby(self.GEOID_TRACT_FIELD_NAME)
.size()
)
self.output_df = (
self.output_df.fillna(0).astype(np.int64).sort_index().reset_index()
)
self.output_df[self.ELIGIBLE_FUDS_BINARY_FIELD_NAME] = np.where(
self.output_df[self.ELIGIBLE_FUDS_COUNT_FIELD_NAME] > 0.0,
True,
False,
)