Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/cfpb/hmda-viz-prototype

Prototype for HMDA visualization
https://github.com/cfpb/hmda-viz-prototype

Last synced: about 1 month ago
JSON representation

Prototype for HMDA visualization

Awesome Lists containing this project

README 

        
# Hmda Viz Prototype



This is a **prototype** for HMDA aggregate and disclosure reports. It is based on the current [FFIEC On-Line Reports](http://www.ffiec.gov/hmda/online_rpts.htm).



Our working prototype can be found running in [gh-pages](http://cfpb.github.io/hmda-viz-prototype/).



This project has 2 parts:

1. the front-end for the reports, including the process of navigating the forms to generate the tables,

2. along with the processing of the data to generate the json files.



![Screenshot](screenshot.png)



## Dependencies



### Front-end



- Ruby - for installing Jekyll

- [Jekyll](http://jekyllrb.com/docs/installation/)



### Processing

 - Python

 - PostgreSQL

 - TBD



## Front-end Installation



To launch the site, enter:



```shell

$ jekyll serve

```



The site should now be live at `http://127.0.0.1:4000/hmda-viz-prototype/`.



We're using [Jekyll's built-in support](http://jekyllrb.com/docs/assets/) for Sass.



## Configuration



### Front-end



See [Jekyll's configuration documentation](http://jekyllrb.com/docs/configuration/) and the [_config.yml](_config.yml) in this repo.



## Usage



Currently only the 2013/alabama path works in the UI. You can then select any MSA/MD and then choose a report. Reports 3-x and 4-x are working.



You should now see an HTML table with options to download the json, csv and print.



![Screenshot](report.png)



## Known issues



We are still in the prototyping phase so there is a lot of work happening on both the front-end and processing sides. One obvious issue is that only certain paths currently work for the forms.



## Getting help



If you have questions, concerns, bug reports, etc, please file an issue in this repository's [Issue Tracker](https://github.com/cfpb/hmda-viz-prototype/issues).



## Getting involved



TBD



----



## Open source licensing info

1. [TERMS](TERMS.md)

2. [LICENSE](LICENSE)

3. [CFPB Source Code Policy](https://github.com/cfpb/source-code-policy/)



----



## Credits and references



1. [FFIEC](http://www.ffiec.gov/HmdaAdWebReport/AggWelcome.aspx)

2. [Federal Register] (http://www.gpo.gov/fdsys/pkg/FR-2004-12-20/pdf/04-27425.pdf)

3. 



## Status

**Time to Market**

* Processing time:

    * Aggregate: 2-3 days processing time

    * Disclosure: 30-45 days processing time (assuming 11-1 processing time can be made to match other reports in the 11 series)

* Time needed to test format, accuracy, footnotes, etc: 30 hours

* Remaining error checking items can be found at (https://drive.google.com/open?id=1g6kkDiD4sxoUd10kmYNsmhTsWRL7iWHbf9GjY9B-FMY&authuser=0)

* Estimated time to write code for National reports: 1 week

* Estimated time to adapt code for running server-side using Pyodbc: 1 week

* Estimated time to write unit testing code: 2 weeks

    * Current completion estimate of unit testing code: 10/140 functions, functions still need comprehensive input lists for testing

    * Nose completion percent: [2%](https://github.com/Kibrael/hmda-viz-prototype/tree/gh-pages/processing/cover/index.html)

* Time to refactor code to allow for creating FI disclosure reports at time of submission: 1 week

* Write scripts to support Jekyll build: 3 days

* Issue 172 - hard-code 'NA' for report B: 2 hours

* Issues 142 and 167 - standardize code and precision for mean and median calculations: 2 days



**Files**

Requirements and analysis are saved at (https://drive.google.com/open?id=0B667BGHwjTznb3F0ZmZabUFoc3M&authuser=0)