.xlabel("\ndate")\
.ylabel("cases\n")\
.annotate("\nBayesian training process on empirical data, with anomalies identified")\
.show()
logger.info("regency-level projections")
regencies = sorted(cases.regency.unique())
regency_cases = cases.groupby(["regency", "confirmed"]).size().sort_index().unstack(fill_value = 0).stack()
migration = np.zeros((len(regencies), len(regencies)))
estimates = []
max_len = 1 + max(map(len, regencies))
with tqdm(regencies) as progress:
for regency in regencies:
progress.set_description(f"{regency :<{max_len}}")
(dates, Rt_pred, Rt_CI_upper, Rt_CI_lower, *_) = analytical_MPVS(regency_cases.loc[regency], CI = CI, smoothing = smoothing, totals=False)
estimates.append((regency, Rt_pred[-1], Rt_CI_lower[-1], Rt_CI_upper[-1], linear_projection(dates, Rt_pred, 7)))
estimates = pd.DataFrame(estimates)
estimates.columns = ["regency", "Rt", "Rt_CI_lower", "Rt_CI_upper", "Rt_proj"]
estimates.set_index("regency", inplace=True)
estimates.to_csv("data/SULSEL_Rt_projections.csv")
print(estimates)
gdf = gpd.read_file("data/gadm36_IDN_shp/gadm36_IDN_2.shp")\
.query("NAME_1 == 'Sulawesi Selatan'")\
.merge(estimates, left_on = "NAME_2", right_on = "regency")
choro = plt.choropleth(gdf, mappable = plt.get_cmap(0.4, 1.4, "viridis"))
for ax in choro.figure.axes[:-1]:
plt.sca(ax)
plt.xlim(left = 119, right = 122)
IDN = SIR("IDN", 267.7e6, dT0 = T_pred[-1], Rt0 = Rt_pred[-1], mobility = 0, random_seed = 0).run(14)
logger.info("province-level projections")
migration = np.zeros((len(provinces), len(provinces)))
estimates = []
max_len = 1 + max(map(len, provinces))
with tqdm(provinces) as progress:
for (province, cases) in province_cases.items():
progress.set_description(f"{province :<{max_len}}")
(dates, Rt_pred, Rt_CI_upper, Rt_CI_lower, *_) = analytical_MPVS(cases, CI = CI, smoothing = smoothing)
apr_idx = np.argmax(dates > "31 Mar, 2020")
may_idx = np.argmax(dates >= "01 May, 2020")
max_idx = np.argmax(Rt_pred[apr_idx:may_idx])
apr_max_idx = apr_idx + max_idx
estimates.append((province, Rt_pred[-1], Rt_CI_lower[-1], Rt_CI_upper[-1], max(0, linear_projection(dates, Rt_pred, window, period = 2*weeks)), Rt_pred[apr_max_idx], Rt_CI_lower[apr_max_idx], Rt_CI_upper[apr_max_idx], dates[apr_max_idx], cases.iloc[-1][0]))
progress.update()
estimates = pd.DataFrame(estimates)
estimates.columns = ["province", "Rt", "Rt_CI_lower", "Rt_CI_upper", "Rt_proj", "Rt_max", "Rt_CI_lower_at_max", "Rt_CI_upper_at_max", "date_at_max_Rt", "total_cases"]
estimates.set_index("province", inplace=True)
estimates.to_csv(data/"IDN_only_apr_Rt_max_filtered.csv")
print(estimates)
# choropleths
logger.info("generating choropleths")
gdf = gpd.read_file("data/gadm36_IDN_shp/gadm36_IDN_1.shp").drop(["GID_0", "NAME_0", "GID_1", "VARNAME_1", "NL_NAME_1", "TYPE_1", "ENGTYPE_1", "CC_1", "HASC_1"], axis = 1)
gdf["NAME_1"] = gdf.NAME_1.str.upper().map(lambda s: replacements.get(s, s))
gdf = gdf.merge(estimates, left_on="NAME_1", right_on="province")
sm = mpl.cm.ScalarMappable(
norm = mpl.colors.Normalize(vmin = 0.9, vmax = 1.4),
])\
.title("\nDKI Jakarta: Daily Cases")\
.xlabel("\ndate")\
.ylabel("cases\n")\
.annotate("\nBayesian training process on empirical data, with anomalies identified")\
.show()
logger.info("district-level projections")
district_cases = dkij.groupby(["district", "date_positiveresult"])["id"].count().sort_index()
districts = dkij.district.unique()
migration = np.zeros((len(districts), len(districts)))
estimates = []
max_len = 1 + max(map(len, districts))
with tqdm(districts) as progress:
for district in districts:
progress.set_description(f"{district :<{max_len}}")
(dates, RR_pred, RR_CI_upper, RR_CI_lower, *_) = analytical_MPVS(district_cases.loc[district], CI = CI, smoothing = smoothing, totals=False)
estimates.append((district, RR_pred[-1], RR_CI_lower[-1], RR_CI_upper[-1], linear_projection(dates, RR_pred, window)))
estimates = pd.DataFrame(estimates)
estimates.columns = ["district", "Rt", "Rt_CI_lower", "Rt_CI_upper", "Rt_proj"]
estimates.set_index("district", inplace=True)
estimates.to_csv(data/"Jakarta_Rt_projections.csv")
print(estimates)
logger.info("generating choropleths")
gdf = gdf.merge(estimates, left_on = "NAME_2", right_on = "district")
plt.choropleth(gdf, lambda row: row["NAME_2"]+"\n")\
.adjust(left = 0.06)\
.title("\nDKI Jakarta: $R_t$ by District")\
.show()
.xlabel("date").ylabel("cases")\
.show()
# now, do district-level estimation
smoothing = 10
state_cases["geo_reported"] = state_cases.geo_reported.str.strip()
district_time_series = state_cases.groupby(
["geo_reported", "date_reported"])["date_reported"].count().sort_index()
migration = np.zeros((len(district_names), len(district_names)))
estimates = []
max_len = 1 + max(map(len, district_names))
with tqdm(etl.normalize(district_names)) as districts:
for district in districts:
districts.set_description(f"{district :<{max_len}}")
try:
(dates, Rt_pred, Rt_CI_upper, Rt_CI_lower,
*_) = analytical_MPVS(district_time_series.loc[district],
CI=CI,
smoothing=convolution(window=smoothing),
totals=False)
estimates.append(
(district, Rt_pred[-1], Rt_CI_lower[-1], Rt_CI_upper[-1],
linear_projection(dates, Rt_pred, smoothing)))
except (IndexError, ValueError):
estimates.append((district, np.nan, np.nan, np.nan, np.nan))
estimates = pd.DataFrame(estimates)
estimates.columns = ["district", "Rt", "Rt_CI_lower", "Rt_CI_upper", "Rt_proj"]
estimates.set_index("district", inplace=True)
estimates.to_csv(data / "Rt_estimates_private.csv")
print(estimates)
_, r = plt.xlim()
plt.xlim(left = pd.Timestamp("August 01, 2020"), right = r)
plt.show()
gdf = gpd.read_file(data/f"{code}.json")
district_names = sorted(gdf.district)
district_time_series = get_time_series(df[df.detected_state == state], "detected_district").Hospitalized
migration = np.zeros((len(district_names), len(district_names)))
estimates = []
max_len = 1 + max(map(len, district_names))
with tqdm(district_time_series.index.get_level_values(0).unique()) as districts:
for district in districts:
districts.set_description(f"{district :<{max_len}}")
try:
(dates, Rt_pred, Rt_CI_upper, Rt_CI_lower, *_) = analytical_MPVS(district_time_series.loc[district], CI = CI, smoothing = notched_smoothing(window = smoothing), totals = False)
estimates.append((district, Rt_pred[-1], Rt_CI_lower[-1], Rt_CI_upper[-1], linear_projection(dates, Rt_pred, smoothing, period = 7)))
except (IndexError, ValueError):
estimates.append((district, np.nan, np.nan, np.nan, np.nan))
estimates = pd.DataFrame(estimates).dropna()
estimates.columns = ["district", "Rt", "Rt_CI_lower", "Rt_CI_upper", "Rt_proj"]
estimates.set_index("district", inplace=True)
estimates = estimates.clip(0)
estimates.to_csv(data/f"Rt_estimates_{code}.csv")
print(estimates)
print(estimates.Rt.max(), estimates.Rt_proj.max())
print(estimates.Rt.min(), estimates.Rt_proj.min())
gdf = gdf.merge(estimates, left_on = "district", right_on = "district")
plt.choropleth(gdf, mappable = plt.get_cmap(*Rt_range))\
.title(f"\n{state}: $R_t$ by District")\
.show()