# Forecast date is equal to the date of the last known datapoint, unless manually supplied
forecast_date = region.index[-1]
region = region[region.index <= forecast_date]
# Early exit: If there are less than DATAPOINT_COUNT_MIN datapoints
if len(region) < DATAPOINT_COUNT_MIN: continue
# Define the subfolder that will hold the output assets
forecast_chart = ROOT / 'output' / 'charts' / ('%s_US_%s.svg' %
(forecast_date, key))
# Perform forecast
forecast_data = forecast(region['Confirmed'], predict_window)
# Output charts as SVG files
plot_forecast(forecast_chart, region['Confirmed'], forecast_data)
# Aggregate forecast data
for idx in forecast_data.index:
forecast_df.loc[(key, idx), 'CountryCode'] = country_code
forecast_df.loc[(key, idx), 'CountryName'] = country_name
forecast_df.loc[(key, idx), 'ForecastDate'] = forecast_date
forecast_df.loc[(key, idx), 'Estimated'] = '%.03f' % forecast_data[idx]
forecast_df.loc[(key, idx),
'ForecastChart'] = forecast_chart.relative_to(ROOT /
'output')
for idx in region['Confirmed'].index:
forecast_df.loc[(key, idx), 'Confirmed'] = int(region.loc[idx,
'Confirmed'])
# Save output to CSV and JSON
if __name__ == "__main__":
df = pd.read_csv("../data/India_OWID_with_mobility_data.csv")
dataset = Data(df, 1, 1)
dataset.smoothen_df()
X, y = dataset.sliding_window(config["features"], dataset.previous_days,
dataset.forecast_days - 1)
print(len(X))
print(X.shape)
print(y.shape)
model = CoronaVirusPredictor(1, 1, 1)
tscv = TimeSeriesSplit()
r2_scores, test_error_mae, test_error_mse, train, test = train_model_with_crossval(
model,
X,
y,
tscv,
)
print(r2_scores)
for i in range(1, 6):
plot_forecast(train["pred"][i], train["true"][i], test["pred"][i],
test["true"][i])
exit(0)
gg, train_hist, test_hist = train_model(model, config, dataset)
plt.plot(train_hist, label="train")
plt.plot(test_hist, label="test")
plt.legend(loc="best")
plt.show()
# Store the charts in a helper JSON file
chart_outputs[key] = {}
confirmed = subset['Confirmed'].dropna()
if len(confirmed) > 0:
plot_column(charts_root / fname_confirmed, confirmed)
chart_outputs[key]['Confirmed'] = fname_confirmed
deaths = subset['Deaths'].dropna()
if len(deaths) > 0:
plot_column(charts_root / fname_deaths, deaths)
chart_outputs[key]['Deaths'] = fname_deaths
try:
# Get the estimated cases from the forecast
estimated = df_forecast[df_forecast['Key'] == key].set_index(
['Date'])['Estimated']
if len(estimated) == 0: continue
# Line up the indices for both the estimated and confirmed cases
estimated = estimated[estimated.index >= confirmed.index[0]].dropna()
confirmed = confirmed[confirmed.index >= estimated.index[0]].dropna()
if len(estimated) > len(confirmed):
plot_forecast(charts_root / fname_forecast, confirmed, estimated)
chart_outputs[key]['Forecast'] = fname_forecast
except Exception as exc:
print('Unexpected error: %r' % sys.exc_info()[0], file=sys.stderr)
print(subset, file=sys.stderr)
# Output the chart map to stdout
json.dump(chart_outputs, sys.stdout)
]).query('~index.duplicated()')
subset = subset.reset_index().set_index('Date').sort_index().iloc[-14:]
fname_forecast = ('%s_%s_forecast.svg' % (prefix, suffix))
fname_confirmed = ('%s_%s_confirmed.svg' % (prefix, suffix))
fname_deaths = ('%s_%s_deaths.svg' % (prefix, suffix))
# Store the charts in a helper JSON file
chart_outputs[suffix] = {}
confirmed = subset['Confirmed'].dropna()
if len(confirmed) > 0:
plot_column(charts_root / fname_confirmed, confirmed)
chart_outputs[suffix]['Confirmed'] = fname_confirmed
deaths = subset['Deaths'].dropna()
if len(deaths) > 0:
plot_column(charts_root / fname_deaths, deaths)
chart_outputs[suffix]['Deaths'] = fname_deaths
try:
if len(subset['Estimated'].dropna()) > len(confirmed):
plot_forecast(charts_root / fname_forecast, subset['Confirmed'],
subset['Estimated'])
chart_outputs[suffix]['Forecast'] = fname_forecast
except Exception as exc:
print('Unexpected error:', sys.exc_info()[0])
print(subset)
with open(charts_root / 'map.json', 'w') as f:
json.dump(chart_outputs, f)