Merge branch 'develop' into issue_119

Author: quinn-dougherty

__init__.py

@@ -1,22 +1,46 @@
 # Contributing
 
-When contributing to this repository, please first discuss the change you wish to make via [issue](https://github.com/codeforphilly/chime/issues), [Slack chat](https://codeforphilly.org/chat/covid19-chime-penn), or any other method with the owners of this repository before making a change.
+Welcome, grab a cup of coffee, and let's roll our sleeves up! By participating in this project, you agree to abide by our [code of conduct](CODE_OF_CONDUCT.md).
 
-Please note we have a [code of conduct](CODE_OF_CONDUCT.md), please follow it in all your interactions with the project.
+## What should I work on?
 
-## Active Needs
+As of March 19, help is especially wanted in these areas:
 
-We maintain a list of [issues in the Code for Philly repo](https://github.com/codeforphilly/chime/issues) that encompasses work requested by stakeholders and technologists.  Each issue is designed to be small enough for one person to implement.
+- **Python wizards** for application development ([#chime-app](https://codeforphilly.org/chat?channel=chime-app))
+  - Hardening the code
+  - Making it more flexible
+  - Unit tests!
+- **DevOps heroes** for application hosting ([#chime-ops](https://codeforphilly.org/chat?channel=chime-ops))
+  - Maintaining Penn's instance
+  - Ensuring easy redeploys
+- **Math and stats friends** for model improvements ([#chime-analysis](https://codeforphilly.org/chat?channel=chime-analysis))
+- **Consulting DevOps champs** to help other public agencies set up their own instances ([#chime-help](https://codeforphilly.org/chat?channel=chime-help)
 
-**Developers** If you are new to the project, check out open issues to see if there is any pre-existing work needed that you can handle. We will need to add your git account to the repo to grant write access - contact us on [Slack](https://codeforphilly.org/chat) in the [covid19-chime-penn](https://codeforphilly.org/chat?channel=covid19-chime-penn) channel.
+## Before You Begin
 
-## Change Management
+- Join the [#covid19-chime-penn](https://codeforphilly.org/chat?channel=covid19-chime-penn) channel, and [request write access](https://codeforphilly.slack.com/archives/CV4NGQYMP/p1584665484368300) to the `CodeforPhilly/chime` Github repository.
+- Our highest-priority work is organized in the [Github Project Management board](https://github.com/CodeForPhilly/chime/projects/2). Look for an issue matching your interests & skills in one of the "Ready" columns.
+- Assign yourself to the issue and add a comment to briefly describe your plan.
+- For new ideas, please first discuss the change(s) you wish to make via [issue](https://github.com/codeforphilly/chime/issues) or appropriate Slack channel (platform/app - #chime-app, devops - #chime-ops, modeling - #chime-analysis). 
 
-- Base all work on the latest `develop` branch available in [Code for Philly's fork of the `chime` repository](https://github.com/CodeForPhilly/chime).
-- Submit your work in the form of a pull request against that same branch.
-- Check in on your pull request periodically after you submit it to see if any changes have been requested or any merge conflicts have arisen from other work getting integrated.
-- If a merge conflict arises, please rebase your branch against the latest `develop` branch and force-push the new branch as early as you can. You may need to do this more than once before your changes get merged. Do your best to keep your branch in a mergeable state until it is finished being reviewed and accepted.
+## Making and Submitting Changes
+
+- [Fork](https://help.github.com/en/github/getting-started-with-github/fork-a-repo) the [CodeforPhilly/chime](https://github.com/CodeForPhilly/chime) repo.
+- Base your work on the `develop` branch.
+- Submit pull requests from your fork, also against the `develop` branch of the `CodeforPhilly/chime` repo.
+- Check your pull request periodically to see if any changes have been requested or any merge conflicts have arisen.
+- If a merge conflict arises, rebase against the latest `develop` branch and force-push the new branch as early as you can. You may need to do this more than once before your changes get merged. Do your best to keep your branch in a mergeable state until it is finished being reviewed and accepted.
+- Note: Frequent contributors with write access can submit pull requests from a new branch in the `CodeforPhilly/chime` repository.
+
+## Review & Release
+
+<!-- Currently establishing & clarifying the release process, this is just a skeleton. -->
+
+- Maintainers will review pull requests, asking for changes as needed.
+- Once a PR has 1 approval and passes automated tests, maintainers will merge to release branches.
+  - Exception: Documentation or infrastructure support PRs can be merged directly to master.
+- Deploys to production must be signed off on by @beckerfluffle or @cchivers.
 
 ## Requesting Functionality
 
-**Stakeholders/Users** Find us on [Slack](https://codeforphilly.org/chat) in the [#covid19-chime-penn](https://codeforphilly.org/chat/covid19-chime-penn) channel to see if your need is already being addressed.
+**Stakeholders/Users**: Check the [#covid19-chime-penn](https://codeforphilly.org/chat/covid19-chime-penn) channel on [Slack](https://codeforphilly.org/chat) to see if your need is already being addressed.

docs/CONTRIBUTING.md

@@ -32,7 +32,7 @@
 
 ## Deploying a New CHIME Version
 
-1. [MaCreateke a new release](release-process.md)
+1. [Create a new release](release-process.md)
 2. Wait for the `Docker` GitHub Actions workflow to complete
 3. Verify the new version shows up at the top as "Latest version" here: https://github.com/CodeForPhilly/chime/packages/155340
 4. Run:

docs/operations/chime-live-cluster.md

@@ -1,9 +1,11 @@
 from typing import Tuple
 
 import numpy as np
+import streamlit as st
 
 
 # The SIR model, one time step
+@st.cache
 def sir(y, beta, gamma, N):
     S, I, R = y
     Sn = (-beta * S * I) + S
@@ -21,6 +23,7 @@ def sir(y, beta, gamma, N):
 
 
 # Run the SIR model forward in time
+@st.cache
 def sim_sir(
     S, I, R, beta, gamma, n_days, beta_decay=0
 ) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
@@ -37,7 +40,7 @@ def sim_sir(
     s, i, r = np.array(s), np.array(i), np.array(r)
     return s, i, r
 
-
+@st.cache
 def get_hospitalizations(
     infected: np.ndarray, rates: Tuple[float, float, float], market_share: float
 ) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:

penn_chime/models.py

@@ -46,7 +46,7 @@ def display_header(
     st.markdown(
         """*This tool was developed by the [Predictive Healthcare team](http://predictivehealthcare.pennmedicine.org/) at
     Penn Medicine. For questions and comments please see our
-    [contact page](http://predictivehealthcare.pennmedicine.org/contact/). Code can be found on [Github](https://github.com/pennsignals/chime).
+    [contact page](http://predictivehealthcare.pennmedicine.org/contact/). Code can be found on [Github](https://github.com/CodeForPhilly/chime).
     Join our [Slack channel](https://codeforphilly.org/chat?channel=covid19-chime-penn) if you would like to get involved!*"""
     )
 

penn_chime/presentation.py

@@ -1,7 +1,8 @@
 import numpy as np
 import pandas as pd
+import streamlit as st
 
-
+@st.cache
 def build_admissions_df(n_days, hosp, icu, vent) -> pd.DataFrame:
     days = np.array(range(0, n_days + 1))
     data_dict = dict(zip(["day", "hosp", "icu", "vent"], [days, hosp, icu, vent]))
@@ -12,7 +13,7 @@ def build_admissions_df(n_days, hosp, icu, vent) -> pd.DataFrame:
     projection_admits["day"] = range(projection_admits.shape[0])
     return projection_admits
 
-
+@st.cache
 def build_census_df(projection_admits, hosp_los, icu_los, vent_los) -> pd.DataFrame:
     """ALOS for each category of COVID-19 case (total guesses)"""
     n_days = np.shape(projection_admits)[0]

penn_chime/utils.py

@@ -1,7 +1,12 @@
 import pytest
+import pandas as pd
 
-from penn_chime.models import sir
-from penn_chime.presentation import display_header
+from app import (S_default, known_infections, known_cases, current_hosp, doubling_time, relative_contact_rate,
+                 hosp_rate, icu_rate, vent_rate, hosp_los, icu_los, vent_los, market_share, S, initial_infections,
+                 detection_prob, hospitalization_rates, I, R, beta, gamma, n_days, beta_decay,
+                 projection_admits, alt)
+from penn_chime.models import sir, sim_sir
+from penn_chime.presentation import display_header, new_admissions_chart
 
 
 # set up
@@ -75,3 +80,52 @@ def test_sir():
         0.20297029702970298,
         0.0049504950495049506,
     ), "This contrived example should work"
+
+
+def test_sim_sir():
+    """
+    Rounding to move fast past decimal place issues
+    """
+    s, i, r = sim_sir(S, I, R, beta, gamma, n_days, beta_decay=beta_decay)
+    assert round(s[0], 0) == 4119405
+    assert round(s[-1], 2) == 3421436.31
+    assert round(i[0], 2) == 533.33
+    assert round(i[-1], 2) == 418157.62
+    assert round(r[0], 0) == 0
+    assert round(r[-1], 2) == 280344.40
+
+
+def test_initial_conditions():
+    """
+    Note: For the rates (ie hosp_rate) - just change the value, leave the "100" alone.
+        Easier to change whole numbers than decimals.
+    """
+    assert current_hosp == known_cases
+    assert doubling_time == 6
+    assert relative_contact_rate == 0
+    assert hosp_rate == 5 / 100
+    assert icu_rate == 2 / 100
+    assert vent_rate == 1 / 100
+    assert hosp_los == 7
+    assert icu_los == 9
+    assert vent_los == 10
+    assert market_share == 15 / 100
+    assert S == S_default
+    assert initial_infections == known_infections
+
+
+def test_derived_variables():
+    assert hospitalization_rates == (hosp_rate, icu_rate, vent_rate)
+    assert detection_prob == initial_infections / (current_hosp / market_share / hosp_rate)
+
+
+def test_new_admissions_chart():
+    chart = new_admissions_chart(alt, projection_admits, n_days - 10)
+    assert type(chart) == alt.Chart
+    assert chart.data.iloc[1].Hospitalized < 1
+    assert round(chart.data.iloc[49].ICU, 0) == 43
+    with pytest.raises(TypeError):
+        new_admissions_chart()
+
+    empty_chart = new_admissions_chart(alt, pd.DataFrame(), -1)
+    assert empty_chart.data.empty