Spaces:
Sleeping
Sleeping
Update main.py
Browse files
main.py
CHANGED
@@ -35,12 +35,13 @@ def geocode_address(address):
|
|
35 |
|
36 |
return lat, lon
|
37 |
|
38 |
-
|
|
|
39 |
|
40 |
start_date = pd.Timestamp(str(start_date)).strftime('%Y%m%d%H')
|
41 |
end_date = pd.Timestamp(str(end_date)).strftime('%Y%m%d%H')
|
42 |
-
date_years = pd.date_range(start=start_date, end=end_date, freq='M')
|
43 |
-
date_range_days = pd.date_range(start_date, end_date)
|
44 |
years = list(set([d.year for d in date_years]))
|
45 |
|
46 |
if len(years) == 0:
|
@@ -49,6 +50,7 @@ def get_hail_data(address, start_date, end_date, radius_miles, get_max):
|
|
49 |
# Geocode Address
|
50 |
lat, lon= geocode_address(address)
|
51 |
|
|
|
52 |
# Convert Lat Lon to row & col on Array
|
53 |
transform = pickle.load(open('Data/hrrr_crs.pkl', 'rb'))
|
54 |
row, col = rasterio.transform.rowcol(transform['affine'], lon, lat)
|
@@ -62,19 +64,20 @@ def get_hail_data(address, start_date, end_date, radius_miles, get_max):
|
|
62 |
|
63 |
files_choosen = [i for i in files if any(i for j in years if str(j) in i)]
|
64 |
|
|
|
65 |
# Query and Collect H5 Data
|
66 |
all_data = []
|
67 |
all_dates = []
|
68 |
for file in files_choosen:
|
69 |
with h5py.File(file, 'r') as f:
|
70 |
# Get Dates from H5
|
71 |
-
dates = f['
|
72 |
date_idx = np.where((dates >= int(start_date))
|
73 |
& (dates <= int(end_date)))[0]
|
74 |
|
75 |
# Select Data by Date and Radius
|
76 |
dates = dates[date_idx]
|
77 |
-
data = f['
|
78 |
radius_miles+1, col-radius_miles:col+radius_miles+1]
|
79 |
|
80 |
all_data.append(data)
|
@@ -95,14 +98,14 @@ def get_hail_data(address, start_date, end_date, radius_miles, get_max):
|
|
95 |
if get_max == True:
|
96 |
data_max = np.max(data_mat, axis=(1, 2))
|
97 |
df_data = pd.DataFrame({'Date': dates_all,
|
98 |
-
'
|
99 |
# Get all Data
|
100 |
else:
|
101 |
data_all = list(data_mat)
|
102 |
df_data = pd.DataFrame({'Date': dates_all,
|
103 |
-
'
|
104 |
|
105 |
-
df_data['Date'] = pd.to_datetime(df_data['Date'], format='%Y%m%d')
|
106 |
df_data = df_data.set_index('Date')
|
107 |
|
108 |
df_data = df_data.reindex(date_range_days, fill_value=0).reset_index().rename(
|
@@ -116,10 +119,10 @@ def get_hail_data(address, start_date, end_date, radius_miles, get_max):
|
|
116 |
async def predict(address: str, start_date: str, end_date: str, radius_miles: int, get_max: bool):
|
117 |
|
118 |
try:
|
119 |
-
results =
|
120 |
end_date, radius_miles, get_max)
|
121 |
except:
|
122 |
-
results = pd.DataFrame({'Date': ['error'], '
|
123 |
|
124 |
return results.to_json()
|
125 |
|
|
|
35 |
|
36 |
return lat, lon
|
37 |
|
38 |
+
|
39 |
+
def get_data(address, start_date, end_date, radius_miles, get_max):
|
40 |
|
41 |
start_date = pd.Timestamp(str(start_date)).strftime('%Y%m%d%H')
|
42 |
end_date = pd.Timestamp(str(end_date)).strftime('%Y%m%d%H')
|
43 |
+
date_years = pd.date_range(start=start_date[:-2], end=end_date[:-2], freq='M')
|
44 |
+
date_range_days = pd.date_range(start_date[:-2], end_date[:-2], freq='H')
|
45 |
years = list(set([d.year for d in date_years]))
|
46 |
|
47 |
if len(years) == 0:
|
|
|
50 |
# Geocode Address
|
51 |
lat, lon= geocode_address(address)
|
52 |
|
53 |
+
|
54 |
# Convert Lat Lon to row & col on Array
|
55 |
transform = pickle.load(open('Data/hrrr_crs.pkl', 'rb'))
|
56 |
row, col = rasterio.transform.rowcol(transform['affine'], lon, lat)
|
|
|
64 |
|
65 |
files_choosen = [i for i in files if any(i for j in years if str(j) in i)]
|
66 |
|
67 |
+
|
68 |
# Query and Collect H5 Data
|
69 |
all_data = []
|
70 |
all_dates = []
|
71 |
for file in files_choosen:
|
72 |
with h5py.File(file, 'r') as f:
|
73 |
# Get Dates from H5
|
74 |
+
dates = f['date_time_hr'][:]
|
75 |
date_idx = np.where((dates >= int(start_date))
|
76 |
& (dates <= int(end_date)))[0]
|
77 |
|
78 |
# Select Data by Date and Radius
|
79 |
dates = dates[date_idx]
|
80 |
+
data = f['APCP'][date_idx, row-radius_miles:row +
|
81 |
radius_miles+1, col-radius_miles:col+radius_miles+1]
|
82 |
|
83 |
all_data.append(data)
|
|
|
98 |
if get_max == True:
|
99 |
data_max = np.max(data_mat, axis=(1, 2))
|
100 |
df_data = pd.DataFrame({'Date': dates_all,
|
101 |
+
'APCP_max': data_max})
|
102 |
# Get all Data
|
103 |
else:
|
104 |
data_all = list(data_mat)
|
105 |
df_data = pd.DataFrame({'Date': dates_all,
|
106 |
+
'APCP_all': data_all})
|
107 |
|
108 |
+
df_data['Date'] = pd.to_datetime(df_data['Date'], format='%Y%m%d%H')
|
109 |
df_data = df_data.set_index('Date')
|
110 |
|
111 |
df_data = df_data.reindex(date_range_days, fill_value=0).reset_index().rename(
|
|
|
119 |
async def predict(address: str, start_date: str, end_date: str, radius_miles: int, get_max: bool):
|
120 |
|
121 |
try:
|
122 |
+
results = get_data(address, start_date,
|
123 |
end_date, radius_miles, get_max)
|
124 |
except:
|
125 |
+
results = pd.DataFrame({'Date': ['error'], 'APCP_max': ['error']})
|
126 |
|
127 |
return results.to_json()
|
128 |
|