VitalSigns 0.0.10

Python Scripts for BNIA-JFIs Vital Signs Data

Vital Signs

Scripts to create our annual, publicly-available, community-focused datasets; for Baltimore City.

Hi! We are BNIA-JFI.

This package was made to help with data handling.

Included

• Functions built and used by BNIA for day to day tasks.
• Made to be shared via IPYNB/ Google Colab notebooks with in-built examples using 100% publicly accessible data & links.
• Online documentation and PyPi libraries created from the notebooks.

TOC

VitalSigns uses functions found in our Dataplay Module and vice-versa.

ACS Scripts

These scripts will download and clean ACS data for Baltimore and then construct indicators from the data.

AcsDownload.ipynb Mount notebook to google drive Read in ACS Meta Data from XLSX and the crosswalk data from a csv Add path to python script that performs download function Enter a year and Run the download function for every record in XLSX sheet For each dataset, remove columnID’s then save it as Raw Then, Append Community Names using a crosswalk and save again in as Clean

AcsIndicators.ipynb Mount notebook to google drive Read in ACS Meta Data from XLSX Prepare the Compare Historic Data

• For Each Indicator
• Grab its Meta Data
• If the Indicator is valid for the year, and uses ACS Data, and method exists
• retrieve the Python ACS indicator
• Put Baltimore City at the bottom of the list
• Write the results back into the XL dataframe
• Save the Dataset
• drop columns with any empty values
• Save the Data xlsx Do comparison to historic year if exists. Write xlsx.

Usage Instructions

Install the Package

The code is on PyPI so you can install the scripts as a python library using the command:

!pip install VitalSigns geopandas

{% include important.html content='Contributers should follow the maintanance instructions and will not need to run this step. ' %}>

Their modules will be retrieved from the VitalSigns-GDrive repo they have mounted into their Colabs Enviornment.

Then...

Import Modules

1. Import the installed module into your code:

from VitalSigns.acsDownload import retrieve_acs_data 

2. use it

retrieve_acs_data(state, county, tract, tableId, year, saveAcs)

Now you could do something like merge it to another dataset!

from dataplay.merge import mergeDatasets
mergeDatasets(left_ds=False, right_ds=False, crosswalk_ds=False,  use_crosswalk = True, left_col=False, right_col=False, crosswalk_left_col = False, crosswalk_right_col = False, merge_how=False, interactive=True)

Getting Help

You can get information on the package by using the help command.

Here we look at the package's modules:

import VitalSigns
help(VitalSigns)

Help on package VitalSigns:

NAME
    VitalSigns

PACKAGE CONTENTS
    BCPSS
    HUD
    _nbdev
    acsDownload
    bidbaltimore
    bpd
    citistat
    cityfinance
    closecrawl
    create
    dhr
    enoch
    fares
    fdic
    indicators
    infousa
    liquor
    mdprop
    rbintel
    tidyaddr
    treebaltimore
    utils

VERSION
    0.0.5

FILE
    /usr/local/lib/python3.7/dist-packages/VitalSigns/__init__.py

Lets take a look at what functions the geoms module provides:

import VitalSigns.acsDownload
help(VitalSigns.acsDownload)

/usr/local/lib/python3.7/dist-packages/VitalSigns/acsDownload.py:27: FutureWarning: Passing a negative integer is deprecated in version 1.0 and will not be supported in future version. Instead, use None to not limit the column width.
  pd.set_option('display.max_colwidth', -1)


Help on module VitalSigns.acsDownload in VitalSigns:

NAME
    VitalSigns.acsDownload - # AUTOGENERATED! DO NOT EDIT! File to edit: notebooks/90_ACS_Explore_and_Download.ipynb (unless otherwise specified).

FUNCTIONS
    retrieve_acs_data(state, county, tract, tableId, year, save)

DATA
    __all__ = ['retrieve_acs_data']
    __warningregistry__ = {'version': 268, ('Passing a negative integer is...

FILE
    /usr/local/lib/python3.7/dist-packages/VitalSigns/acsDownload.py




/usr/local/lib/python3.7/dist-packages/psycopg2/__init__.py:144: UserWarning: The psycopg2 wheel package will be renamed from release 2.8; in order to keep installing from binary please use "pip install psycopg2-binary" instead. For details see: <http://initd.org/psycopg/docs/install.html#binary-install-from-pypi>.
  """)

And here we can look at an individual function and what it expects:

import VitalSigns.acsDownload
help(VitalSigns.acsDownload.retrieve_acs_data)

Help on function retrieve_acs_data in module VitalSigns.acsDownload:

retrieve_acs_data(state, county, tract, tableId, year, save)

Examples

So heres an example:

Import your modules

from VitalSigns.acsDownload import retrieve_acs_data

Read in some data

Define our download parameters.

More information on these parameters can be found in the tutorials!

# Our download function will use Baltimore City's tract, county and state as internal paramters
# Change these values in the cell below using different geographic reference codes will change those parameters
tract = '*'
county = '510'
state = '24'

# Specify the download parameters the function will receieve here
tableId = 'B19001'
year = '17'
saveAcs = False

And download the Baltimore City ACS data using the imported VitalSigns library.

df = retrieve_acs_data(state, county, tract, tableId, year, saveAcs)
df.head(1)

Number of Columns 17

<style scoped> .dataframe tbody tr th:only-of-type { vertical-align: middle; }

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

	B19001_001E_Total	B19001_002E_Total_Less_than_$10,000	B19001_003E_Total_$10,000_to_$14,999	...	state	county	tract
NAME
Census Tract 2710.02	1510	209	73	...	24	510	271002

1 rows × 20 columns

From there, you can go on to do even greater things using our dataplay library. Like these visuals:

Have Fun!

MISC

This section is not definite but provides a good idea of how our scripts are made.

Basic Indicator Creation Outline

• ? Count = 1
• Create the num and denom
• filter num denom
• ? sum/ median = ungrouped.median
• group by csa
• ? bcity = median or sum
• perform the calculation
• compare years

Miscellaneous things I should have for every notebook

• Module/filenames need to be fixed.
• RB Intel has the best prelim analysis script. The others are messed up a bit?
• include links indicator Esri and Bnia pages details on category, name, description, years
• Don’t drop columns at end, but keep selected at beginning.
• Merge on CSA for ordering
• Bcity Median gets calculated before aggregation. Appended after
• Add Years in header. Use denom and numerator as var names.
• Code to compare past years

FOR CONTRIBUTERS

Dev Instructions

From a local copy of the git repo: 0. Clone the repo local onto GDrive

• Via Direct-DL&Drive-Upload or Colab/Terminal/Git
• git clone https://github.com/BNIA/VitalSigns.git

1. Update the the IPYNB

• From the GDrive VitalSigns folder via Colabs

2. Build the new libraries from these NBs

• Using this index.ipynb
• • Mount the Colabs Enviornment (and navigate to) the GDrive VitalSigns folder
• • run !nbdev_build_lib to build .py modules.

3. Test the Library/ Modules

• Using the same runtime as step 2's index.ipynb.
• • Do not install the module from PyPi (if published) and then...
• • Import your module ( from your VitalSigns/VitalSigns)
• • If everything runs properly, go to step 5.

4. Edit modules directly

• Within the same runtime as step 2/3's index.ipynb...
• • Locate the VitalSigns/VitalSigns using the colab file nav
• • double-click the .py modules in the file nav to open them in an in-browser editor
• Make changes and return to step 3 with the following caveat:
• • Use the hot module reloading to ensure updates are auto-re-imported
• • %load_ext autoreload %autoreload 2
• Then when finished, persist the changes from the .py modules back to the .ipynb docs
• • via !nbdev_update_lib and !relimport2name

5. Create Docs, Push to Github, and Publish to PyPI

• All done via nbdev
• Find more notes I made on that here: dataplay > nbdev notes
• !nbdev_build_docs --force_all True --mk_readme True
• !git commit -m ...
• %%capture ! pip install twine
• !nbdev_bump_version
• ! make pypi

# https://nbdev.fast.ai/tutorial.html#Set-up-prerequisites
# settings.ini > requirements = fastcore>=1.0.5 torchvision<0.7
# https://nbdev.fast.ai/tutorial.html#View-docs-locally
# console_scripts = nbdev_build_lib=nbdev.cli:nbdev_build_lib
# https://nbdev.fast.ai/search

Dev Scripts