def setUpClass(cls):
cls.n = 20001
cls.seed = 1001
# change this to True if you need to re-generate the baseline
cls.generateBaseline = regenerate
cls.pdfDir = sc.thisdir(__file__, "regression", "report")
cls.expectedDir = sc.thisdir(__file__, "regression", "expected")
cls.datadir = sc.thisdir(__file__, os.pardir, "data")
shutil.rmtree(cls.pdfDir, ignore_errors=True)
os.makedirs(cls.pdfDir, exist_ok=True)
def test_api(do_plot=False):
# Read data
filename = sc.thisdir(__file__, '..', 'examples', 'du_fig1a_annotated.png')
xlim = [-10, 20]
ylim = [0, 15]
data = dt.datathief(filename, xlim=xlim, ylim=ylim, debug=True)
# Plotting
if do_plot:
import pylab as pl
pl.figure(figsize=(12, 4))
ax1 = pl.subplot(1, 2, 1)
ax1.imshow(pl.imread(filename))
ax1.set_title('Original')
ax2 = pl.subplot(1, 2, 2)
ax2.bar(data.x, data.y)
ax2.set_title('Extracted')
ax2.set_xlabel('Serial interval')
ax2.set_ylabel('Frequency')
pl.show()
return data
def load_fonts(folder=None, rebuild=False, verbose=False, **kwargs):
'''
Helper function to load custom fonts for plotting -- (usually) not for the user.
Note: if fonts don't load, try running ``cv.settings.load_fonts(rebuild=True)``,
and/or rebooting the system.
Args:
folder (str): the folder to add fonts from
rebuild (bool): whether to rebuild the font cache
verbose (bool): whether to print out progress/errors
'''
if folder is None:
folder = str(sc.thisdir(__file__, aspath=True) / 'data' / 'assets')
sc.fonts(add=folder, rebuild=rebuild, verbose=verbose, **kwargs)
# Try to find the font, and if it succeeds, update the styles
try:
name = 'Mulish'
fm.findfont(name, fallback_to_default=False
) # Raise an exception if the font isn't found
rc_simple['font.family'] = name # Need to set both
rc_covasim['font.family'] = name
if verbose: print(f'Default Covasim font reset to "{name}"')
except Exception as E:
if verbose: print(f'Could not find font {name}: {str(E)}')
return
def test_thisdir():
sc.heading('Test getting the current file directory')
import os
thisdir = sc.thisdir(__file__)
assert os.path.split(thisdir)[-1] == 'tests'
print(f'Current folder: {thisdir}')
return thisdir
def test_data():
sc.heading('Data loading')
data = cv.load_data(os.path.join(sc.thisdir(__file__), 'example_data.csv'))
sc.pp(data)
# Check that it is looking for the right file
with pytest.raises(FileNotFoundError):
data = cv.load_data(datafile='file_not_found.csv')
return data
def test_baseline():
''' Compare the current default sim against the saved baseline '''
# Load existing baseline
filepath = sc.makefilepath(filename=baseline_filename,
folder=sc.thisdir(__file__))
baseline = sc.loadjson(filepath)
old = baseline[baseline_key]
# Calculate new baseline
sim = cv.Sim(verbose=0)
sim.run()
new = sim.summary
# Compare keys
errormsg = ''
old_keys = set(old.keys())
new_keys = set(new.keys())
if old_keys != new_keys:
errormsg = f"Keys don't match!\n"
missing = old_keys - new_keys
extra = new_keys - old_keys
if missing:
errormsg += f' Missing old keys: {missing}\n'
if extra:
errormsg += f' Extra new keys: {extra}\n'
mismatches = {}
for key in old_keys.union(new_keys):
old_val = old[key] if key in old else 'not present'
new_val = new[key] if key in new else 'not present'
if old_val != new_val:
mismatches[key] = {'old': old_val, 'new': new_val}
if len(mismatches):
errormsg = '\nThe following values have changed between old and new!\n'
errormsg += 'Please rerun "tests/unittests/update_baseline" if this is intentional.\n'
errormsg += 'Mismatches:\n'
space = ' ' * 17
for mkey, mval in mismatches.items():
errormsg += f' {mkey}:\n'
errormsg += f'{space}old = {mval["old"]}\n'
errormsg += f'{space}new = {mval["new"]}\n'
# Raise an error if mismatches were found
if errormsg:
prefix = '\nThe following values have changed between the previous baseline and now!\n'
prefix += 'If this is intentional, please rerun "update_baseline" and commit.\n\n'
err = prefix + errormsg
raise ValueError(err)
else:
print('Baseline matches')
return new
def get_version_pars(version, verbose=True):
'''
Function for loading parameters from the specified version.
Args:
version (str): the version to load parameters from
Returns:
Dictionary of parameters from that version
'''
# Define mappings for available sets of parameters -- note that this must be manually updated from the changelog
match_map = {
'0.30.4': ['0.30.4'],
'0.31.0': ['0.31.0'],
'0.32.0': ['0.32.0'],
'1.0.0': ['1.0.0'],
'1.0.1': [f'1.0.{i}' for i in range(1, 4)],
'1.1.0': ['1.1.0'],
'1.1.1': [f'1.1.{i}' for i in range(1, 3)],
'1.1.3': [f'1.1.{i}' for i in range(3, 8)],
'1.2.0': [f'1.2.{i}' for i in range(4)],
'1.3.0': [f'1.3.{i}' for i in range(6)],
'1.4.0': [f'1.4.{i}' for i in range(9)],
'1.5.0': [f'1.5.{i}' for i in range(4)] +
[f'1.6.{i}' for i in range(2)] + [f'1.7.{i}' for i in range(7)],
'2.0.0': [f'2.0.{i}' for i in range(5)] + ['2.1.0'],
'2.1.1': [f'2.1.{i}' for i in range(1, 3)],
'3.0.0': ['3.0.0'],
}
# Find and check the match
match = None
for ver, verlist in match_map.items():
if version in verlist:
match = ver
break
if match is None: # pragma: no cover
options = '\n'.join(sum(match_map.values(), []))
errormsg = f'Could not find version "{version}" among options:\n{options}'
raise ValueError(errormsg)
# Load the parameters
filename = f'pars_v{match}.json'
regression_folder = sc.thisdir(__file__, 'regression')
pars = sc.loadjson(filename=filename, folder=regression_folder)
if verbose:
print(f'Loaded parameters from {match}')
return pars
def get_version_pars(version, verbose=True):
'''
Function for loading parameters from the specified version.
Parameters will be loaded for Covasim 'as at' the requested version i.e. the
most recent set of parameters that is <= the requested version. Available
parameter values are stored in the regression folder. If parameters are available
for versions 1.3, and 1.4, then this function will return the following
- If parameters for version '1.3' are requested, parameters will be returned from '1.3'
- If parameters for version '1.3.5' are requested, parameters will be returned from '1.3', since
Covasim at version 1.3.5 would have been using the parameters defined at version 1.3.
- If parameters for version '1.4' are requested, parameters will be returned from '1.4'
Args:
version (str): the version to load parameters from
Returns:
Dictionary of parameters from that version
'''
# Construct a sorted list of available parameters based on the files in the regression folder
regression_folder = sc.thisdir(__file__, 'regression', aspath=True)
available_versions = [
x.stem.replace('pars_v', '') for x in regression_folder.iterdir()
if x.suffix == '.json'
]
available_versions = sorted(available_versions, key=LooseVersion)
# Find the highest parameter version that is <= the requested version
version_comparison = [
sc.compareversions(version, v) >= 0 for v in available_versions
]
try:
target_version = available_versions[sc.findlast(version_comparison)]
except IndexError:
errormsg = f"Could not find a parameter version that was less than or equal to '{version}'. Available versions are {available_versions}"
raise ValueError(errormsg)
# Load the parameters
pars = sc.loadjson(filename=regression_folder /
f'pars_v{target_version}.json',
folder=regression_folder)
if verbose:
print(f'Loaded parameters from {target_version}')
return pars
def test_sim_data(do_plot=False, do_show=False):
sc.heading('Data test')
pars = dict(
n=2000,
start_day = '2020-01-01',
)
# Create and run the simulation
sim = cv.Sim(pars=pars, datafile=os.path.join(sc.thisdir(__file__), 'example_data.csv'))
sim.run()
# Optionally plot
if do_plot:
sim.plot(do_show=do_show)
return sim
def output(self):
self.log.info(f"Final columns: {', '.join(self.df.columns)}")
self.log.info("First rows of data:")
self.log.info(self.df.head())
here = sc.thisdir(__file__)
data_home = os.path.join(here, self.output_folder)
for g in self.grouping:
key_value = g[0]
filename = f'{sc.sanitizefilename(key_value)}.csv'
filepath = sc.makefilepath(filename=filename, folder=data_home)
self.log.info(f'Creating {filepath}')
mini_df = self.df[self.df.key == key_value]
mini_df.to_csv(filepath)
self.log.info(
f"There are {len(self.grouping)} entities in this dataset.")
self.log.info(f"Saved {len(self.df)} records.")
def get_version_pars(version, verbose=True):
'''
Function for loading parameters from the specified version.
Args:
version (str): the version to load parameters from
Returns:
Dictionary of parameters from that version
'''
regression_folder = sc.thisdir(__file__, 'regression')
pattern = 'pars_v*.json'
requested = pattern.replace('*', version)
filepaths = sc.getfilelist(regression_folder, pattern=pattern)
files = [os.path.basename(f) for f in filepaths]
if requested in files: # If there's an exact match
match = requested
else: # No match, find the nearest matching file
withmatch = files + [requested] # Add this version
withmatch.sort() # Sort the files
index = withmatch.index(requested)
if index > 0:
match = withmatch[
index -
1] # Get latest earlier version -- note that this assumes versions are in alphabetical order, which they currently are!
else:
filestr = '\n'.join(files)
errormsg = f'Could not find version {version} among options:\n{filestr}'
raise ValueError(errormsg)
# Load the parameters
pars = sc.loadjson(filename=match, folder=regression_folder)
if verbose:
if match == requested:
print(f'Loaded parameters from {match}')
else:
print(
f'No exact match for parameters "{version}" found; using "{match}" instead'
)
return pars
def setUpClass(cls) -> None:
# for Testconvert_df_to_json_array()
cls.pars['filepath'] = os.path.join(
sc.thisdir(__file__, 'testdata'),
'Washington_age_bracket_distr_16.dat')
import covasim as cv
import sciris as sc
cv.check_version('0.27.9')
do_plot = 1
datafile = sc.thisdir(__file__, '../../example_data.csv')
pars = sc.objdict(diag_factor=1.0, n_days=30, pop_infected=100)
sim = cv.Sim(pars, datafile=datafile)
case = 2
testprobdict = {
0:
cv.test_prob(symp_prob=0,
asymp_prob=0,
test_sensitivity=1.0,
loss_prob=0.0,
test_delay=0,
start_day=0), # works, no diagnoses
1:
cv.test_prob(symp_prob=1,
asymp_prob=0,
test_sensitivity=1.0,
loss_prob=0.0,
test_delay=0,
start_day=0), # works, ~half diagnosed
2:
cv.test_prob(
)
fig.update_layout(
plot_bgcolor='#fff'
)
fig.update_layout(title={'text': 'Epidemic over time'})
output = {'json': fig.to_json(), 'id': str(sc.uuid())}
d = json.loads(output['json'])
d['config'] = {'responsive': True}
output['json'] = json.dumps(d)
return output
#%% Run the server using Flask
if __name__ == "__main__":
os.chdir(sc.thisdir(__file__))
if len(sys.argv) > 1:
app.config['SERVER_PORT'] = int(sys.argv[1])
else:
app.config['SERVER_PORT'] = 8188
if len(sys.argv) > 2:
autoreload = int(sys.argv[2])
else:
autoreload = 1
app.run(autoreload=autoreload)
import sciris as sc
import logging
from . import version as spv
__all__ = [
'logger', 'checkmem', 'datadir', 'localdatadir', 'version_info',
'set_datadir', 'set_nbrackets', 'validate'
]
# Declaring this here makes it globally available as synthpops.datadir
datadir = None
localdatadir = None
full_data_available = False # this is likely not necesary anymore
# Set the local data folder
thisdir = sc.thisdir(__file__)
localdatadir = os.path.join(thisdir, os.pardir, 'data')
# Replace with local data dir if Dropbox folder is not found
if datadir is None:
full_data_available = True
datadir = localdatadir
# Number of census age brackets to use
nbrackets = [
16, 20
][1] # Choose how many age bins to use -- 20 is only partially supported
matrix_size = 16 # The dimensions of the mixing matrices -- currently only 16 is available
#%% Logger
print('Note -- these tests currently deprecated')
import numpy as np
import pandas as pd
import sciris as sc
import synthpops as sp
import os
import unittest
sp.settings.datadir = sc.thisdir(__file__, 'unittests')
# for Testconvert_df_to_json_array()
pars = sc.objdict(
location_name = 'usa-Washington', # name of the location
property_name = 'population_age_distributions', # name of the property to compare to
filepath = os.path.join(sp.settings.datadir,
'Washington_age_bracket_distr_16.dat'), # path to the file to convert to array
cols_ind = [], # list of column indices to include in array in conversion
int_cols_ind = [], # list of column induces to convert to ints
)
class TestLocation(unittest.TestCase):
"""
These tests need to be run from the root synthpops/tests folder, because some of the tests involve relative
filepath assumptions based on that.
"""
def minimal_test_str(self):
test_str = """{
"location_name": "test_location",
'''
Run examples/*.py using pytest
'''
import pylab as pl
import sciris as sc
from pathlib import Path
import importlib.util as iu
pl.switch_backend(
'agg'
) # To avoid graphs from appearing -- if you want them, run the examples directly
cwd = Path(sc.thisdir(__file__))
examples_dir = cwd.joinpath('../examples')
def run_example(name):
'''
Execute an example py script as __main__
Args:
name (str): the filename without the .py extension
'''
spec = iu.spec_from_file_location("__main__",
examples_dir.joinpath(f"{name}.py"))
module = iu.module_from_spec(spec)
spec.loader.exec_module(module)
def test_run_scenarios():
run_example("run_scenarios")
'''
Tests for things that are not tested in other files, typically because they are
corner cases or otherwise not part of major workflows.
'''
#%% Imports and settings
import os
import pytest
import numpy as np
import sciris as sc
import covasim as cv
do_plot = 1
verbose = -1
debug = 1 # This runs without parallelization; faster with pytest
csv_file = os.path.join(sc.thisdir(), 'example_data.csv')
xlsx_file = os.path.join(sc.thisdir(), 'example_data.xlsx')
cv.options.set(interactive=False) # Assume not running interactively
def remove_files(*args):
''' Remove files that were created '''
for path in args:
if os.path.exists(path):
print(f'Removing {path}')
os.remove(path)
return
#%% Define the tests
"""
Compare current results to baseline
"""
import sciris as sc
import covasim as cv
do_save = True
baseline_filename = sc.thisdir(__file__, 'baseline.json')
benchmark_filename = sc.thisdir(__file__, 'benchmark.json')
baseline_key = 'summary'
def save_baseline(do_save=do_save):
''' Refresh the baseline results '''
print('Updating baseline values...')
sim = cv.Sim(verbose=0)
sim.run()
sim.to_json(filename=baseline_filename, keys=baseline_key)
print('Done.')
return sim
def test_baseline():
''' Compare the current default sim against the saved baseline '''
# Load existing baseline
filepath = sc.makefilepath(filename=baseline_filename,
'''
Converts Excel files to binary for faster loading.
'''
import os
import numpy as np
import pandas as pd
import sciris as sc
folder = os.path.join(sc.thisdir(__file__), os.pardir, 'data', 'demographics',
'contact_matrices_152_countries')
ext1 = '_1.xlsx'
ext2 = '_2.xlsx'
files1 = sc.getfilelist(folder, pattern=f'*{ext1}')
files2 = sc.getfilelist(folder, pattern=f'*{ext2}')
basefilenames = [f[len(folder) + 1:-len(ext1)]
for f in files1] # 7 is the length of the extension
datadict = {}
for fn in basefilenames:
datadict[fn] = {}
for fn in basefilenames:
for i, ext in enumerate([ext1, ext2]):
thisfile = folder + os.sep + fn + ext
print(f'Working on {thisfile}...')
xls = pd.ExcelFile(thisfile)
for sheet_name in xls.sheet_names:
if i == 0:
header = 0
''' Ensure that the right paths can be found ''' import sys import sciris as sc sys.path.append(sc.thisdir(__file__))
def setUpClass(cls) -> None:
cls.example_path = Path(sc.thisdir(__file__)).joinpath('../examples')
cls.resultdir = os.path.join(os.path.dirname(__file__), 'result')
os.makedirs(cls.resultdir, exist_ok=True)
#!/usr/bin/env python3
'''
Run the non-tutorial examples using pytest
'''
import pylab as pl
import sciris as sc
from pathlib import Path
import importlib.util as iu
pl.switch_backend(
'agg'
) # To avoid graphs from appearing -- if you want them, run the examples directly
examples_dir = Path(sc.thisdir(__file__))
def run_example(name):
'''
Execute an example py script as __main__
Args:
name (str): the filename
'''
spec = iu.spec_from_file_location("__main__", examples_dir.joinpath(name))
module = iu.module_from_spec(spec)
spec.loader.exec_module(module)
def test_run_scenarios():
run_example("run_scenarios.py")
# Select and rename columns
target_columns = [
'Country or area', 'Reference date (dd/mm/yyyy)',
'Average household size (number of members)'
]
df = df_raw[target_columns].copy()
df.columns = ['country', 'date', 'size']
# Convert date column to datetime type and replace nodata with NA.
df['date'] = df['date'].apply(lambda d: pd.to_datetime(d, format='%d/%m/%Y'))
df['size'] = df['size'].apply(lambda s: np.nan
if isinstance(s, str) and s == '..' else s)
df = df.dropna()
# Take the most recent household size for each country.
df = df.sort_values(by=['country', 'date']).groupby(by=['country']).last()[[
'size'
]]
un_country_households_dict = df.to_dict()['size'].copy()
# Add to the dictionary commonly used aliases like USA for United States of America.
us_countries = [k.lower() for k in un_country_households_dict]
country_mappings_dict = cv.data.loaders.get_country_aliases()
for alias, name in country_mappings_dict.items():
if name.lower() in us_countries:
un_country_households_dict[alias] = un_country_households_dict[name]
pathname = str(
sc.thisdir(cv.__file__, aspath=True) / 'data' / 'household_size_data.py')
print(f'To update, copy the data below into the following file:\n{pathname}\n')
print(un_country_households_dict)
import sys
import shutil
import fnmatch
import unittest
import tempfile
import sciris as sc
import synthpops as sp
import numpy as np
from examples import plot_age_mixing_matrices
try:
from fpdf import FPDF
except Exception as E:
print(f'Note: could not import fpdf, report not available ({E})')
# import utilities from test directory
testdir = sc.thisdir(__file__, os.pardir)
sys.path.append(testdir)
import utilities
# Whether to remove temporary files generated in the process
remove_files = True
# Whether to regenerate files
regenerate = False
class TestRegression(unittest.TestCase):
@classmethod
def setUpClass(cls):
cls.n = 20001
cls.seed = 1001
def setUpClass(cls) -> None:
sp.settings.datadir = sc.thisdir(__file__, 'testdata')
"""
Compare current results to baseline
"""
import numpy as np
import pandas as pd
import sciris as sc
import covasim as cv
do_save = False
baseline_filename = sc.thisdir(__file__, 'baseline.json')
benchmark_filename = sc.thisdir(__file__, 'benchmark.json')
parameters_filename = sc.thisdir(__file__, 'regression', f'parameters_v{cv.__version__}.json')
baseline_key = 'summary'
def save_baseline(do_save=do_save):
''' Refresh the baseline results '''
print('Updating baseline values...')
sim = cv.Sim(verbose=0)
sim.run()
if do_save:
sim.to_json(filename=baseline_filename, keys=baseline_key)
sim.export_pars(filename=parameters_filename)
print('Done.')
return sim
# Use internal data to create dates; only keep `date` column
df['date'] = pd.to_datetime(df[['year', 'month', 'day']])
df.drop(['dateRep', 'day', 'month', 'year'], inplace=True, axis=1)
# Just to be safe, let's sort by name and date. Probably not
# necessary but better safe than sorry!.
df = df.sort_values(['countriesAndTerritories', 'date'])
# Each data set has a unique name. Let's create groups.
g = df.groupby('countriesAndTerritories')
# The parameter 'day' is the number of days since the first
# day of data collection for the group.
df['day'] = g['date'].transform(lambda x: (x - min(x))).apply(lambda x: x.days)
# We'll 'rename' some of the columns to be consistent
# with the parameters file.
df['new_positives'] = df.cases
df['new_death'] = df.deaths
df['population'] = df.popData2018
df.drop(['cases', 'deaths', 'popData2018'], inplace=True, axis=1)
# And save it to the data directory.
here = sc.thisdir(__file__)
data_home = os.path.join(here, subfolder)
filepath = sc.makefilepath(filename=outputfile, folder=data_home)
log.info(f"Saving to {filepath}")
df.to_csv(filepath)
log.info(f"Script complete")
'''
Confirm that with default settings, all analyzers can be exported as JSONs.
'''
import sciris as sc
import covasim as cv
datafile = sc.thisdir(__file__, aspath=True).parent / 'example_data.csv'
# Create and runt he sim
sim = cv.Sim(analyzers=[
cv.snapshot(days='2020-04-04'),
cv.age_histogram(),
cv.daily_age_stats(),
cv.daily_stats()
],
datafile=datafile)
sim.run()
# Compute extra analyzers
tt = sim.make_transtree()
fit = sim.compute_fit()
# Construct list of all analyzers
analyzers = sim['analyzers'] + [tt, fit]
# Make jsons
jsons = {}
for an in analyzers:
print(f'Working on analyzer {an.label}...')
jsons[an.label] = an.to_json()