search

abtassociates / eva

Eva is a HUD application to aid HMIS Leads with data analysis. It is an open-source project intended for local use by HMIS Administrators in Continuums of Care (CoCs) around the U.S. and its territories.
GNU Affero General Public License v3.0
14 stars 4 forks source link

Fixing all served #542

Closed alex-silverman closed 2 months ago

alex-silverman commented 2 months ago

Had to rework code to handle the fact that some of the selected comparison "variables" are actually comprised of multiple binary variables (e.g. Gender).

alex-silverman commented 2 months ago

I may not have pushed my fix for that. When did you last pull? Have you repulled jic?

On Sun, Aug 11, 2024, 9:04 PM Genelle Denzin @.***> wrote:

@.**** requested changes on this pull request.

hey I didn't have a chance to troubleshoot it but whenever I run Eva, upload the sample data, and select "Gender" and "Age" I get an error that says "object vars2 not found"

— Reply to this email directly, view it on GitHub https://github.com/abtassociates/eva/pull/542#pullrequestreview-2231936603, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAZCFWPHLBN3PSQT5V6SMMTZRACZZAVCNFSM6AAAAABMIZUKW6VHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZDEMZRHEZTMNRQGM . You are receiving this because you authored the thread.Message ID: @.***>

alex-silverman commented 2 months ago

@kiadso weird, but I did see that vars2 issue and fixed it.

alex-silverman commented 2 months ago

@kiadso - I just pushed a number of changes to make sure things are working better and code is cleaner.

I am noticing that the Hispanic-Focus Races/Ethnicities (when methodology type = Inclusive) + Gender is generating an error. I'll work on that tomorrow

alex-silverman commented 2 months ago

@kiadso - okay can you try again? I've just fixed Hispanic-Focused Races/Ethnicities.

alex-silverman commented 2 months ago
  • I'm not seeing the DV Fleeing category on the chart but I verified there are clients that should have that as their DomesticViolenceCategory.
  • Also if you select Age and Gender, the Genders don't make sense. Like there's hardly any cis men/boys at all and there are a lot of "gender expansive, not transgender" clients.
  • When I select Homelessness Type and any other selection, I don't see anything.
  • When I select Age and Race/Ethnicities, the numbers also seem incorrect, like the highest numbers are in Asian/Asian American Alone and Native Hawaiian/Pacific Islander.
  • the numeric labels should use scales::comma(value) so they show with commas
  • I really like the way the drop lists in the top section gray out when you go to this tab, it makes sense
  • if I go to methodology 2 and select Hispanic-Focused Race/Ethnicities + Veteran Status or Veteran Status + All Races/Ethnicities, I get an error: [object Object] thanks for all your work on this!

thanks for these detailed notes. will take a look.

alex-silverman commented 2 months ago
  • I'm not seeing the DV Fleeing category on the chart but I verified there are clients that should have that as their DomesticViolenceCategory.
  • Also if you select Age and Gender, the Genders don't make sense. Like there's hardly any cis men/boys at all and there are a lot of "gender expansive, not transgender" clients.
  • When I select Homelessness Type and any other selection, I don't see anything.
  • When I select Age and Race/Ethnicities, the numbers also seem incorrect, like the highest numbers are in Asian/Asian American Alone and Native Hawaiian/Pacific Islander.
  • the numeric labels should use scales::comma(value) so they show with commas
  • I really like the way the drop lists in the top section gray out when you go to this tab, it makes sense
  • if I go to methodology 2 and select Hispanic-Focused Race/Ethnicities + Veteran Status or Veteran Status + All Races/Ethnicities, I get an error: [object Object] thanks for all your work on this!

Re #1, can you check again? I'm seeing The Currently Fleeing DV category: image

Re #2 and #4, looks like the data became misaligned with the labels. Fixed!

Re #5, added commas!

Re #7, fixed this issue. Weirdly, it had to do with how we force the Race/Ethnciity selection to be the y-label axis. It's weird that the issue didn't show up before.

kiadso commented 2 months ago

this is what I'm seeing for DV and Age (Not DV and Gender): image taking a look at the rest shortly!

kiadso commented 2 months ago

Also I'm noticing the Gender and DV selections seem off as well, it has 53% of those who experienced DV as men, 21% as women- are we sure that's right?

alex-silverman commented 2 months ago

this is what I'm seeing for DV and Age (Not DV and Gender): image taking a look at the rest shortly!

not able to replicate with Demo mode or loading hashed-current-good. Which file are you using?

alex-silverman commented 2 months ago

@kiadso - ready for another review! I've incorporated fixes to the issues you mentioned. The data look more reasonable to me now. If you notice anything, please include the file used to test so I can replicate.

alex-silverman commented 2 months ago
  • Whenever I go to Veteran Status and anything else (regardless of methodology), I don't see the Veteran cells with any numbers.
  • in the detail box, it's showing Household Type and Level of Detail, which they may want to show conditionally
  • Even with these issues, I'm not against merging them in so you can continue to work out the kinks. Let me know if you want to get this all the way there before merging or if you want to go ahead and merge and address any other issues in subsequent branches
alex-silverman commented 2 months ago

@kiadso and @alex-silverman decided to merge this and fix any issues as hotfix branches