Chg numpy.asscalar() calls to numpy.ndarray.item()

re: Deprecation of numpy.asscalar() in numpy=1.16 and its removal in numpy=1.23.

SalishSeaTools/salishsea_tools/geo_tools.py

@@ -212,7 +212,7 @@ def find_closest_model_point(
         else:
             return np.nan, np.nan
     try:
-        j, i = map(np.asscalar, (j_list, i_list))
+        j, i = map(np.ndarray.item, (j_list, i_list))
     except ValueError:
         # Several points within tolerance
         # Calculate distances for all and choose the closest
@@ -226,7 +226,7 @@ def find_closest_model_point(
             np.array([lon] * i_list.size), np.array([lat] * j_list.size),
             lons, lats)
         n = dists.argmin()
-        j, i = map(np.asscalar, (j_list[n], i_list[n]))
+        j, i = j_list.item(n), i_list.item(n)
 
     # If point is on land and land mask is provided
     # try to find closest water point
@@ -244,11 +244,11 @@ def find_closest_model_point(
         else:
             return _spiral_search_for_closest_water_point(
                 j, i, land_mask, lon, lat, model_lons, model_lats)
-    except ValueError: 
+    except ValueError:
         if raiseOutOfBounds:
             raise ValueError(
                 'lat/lon on land and no nearby water point found')
-        else: 
+        else:
             return np.nan, np.nan
 
 def closestPointArray(lons,lats,

SalishSeaTools/salishsea_tools/nc_tools.py

@@ -267,7 +267,7 @@ def timestamp(dataset, tindex, time_var='time_counter'):
     results = []
     for i in tindex:
         try:
-            results.append(time_orig + timedelta(seconds=np.asscalar(time_counter[i])))
+            results.append(time_orig + timedelta(seconds=time_counter[i].item()))
         except IndexError:
             raise IndexError(
                 'time_counter variable has no tindex={}'.format(tindex))

SalishSeaTools/salishsea_tools/wind_tools.py

@@ -255,8 +255,7 @@ def calc_wind_avg_at_point(date_time, weather_path, windji, avg_hrs=-4):
         wind_u = np.concatenate((wind_prev_day.u, wind_u))
         wind_v = np.concatenate((wind_prev_day.v, wind_v))
         wind_t = np.concatenate((wind_prev_day.time, wind_t))
-    i_date_time = np.asscalar(
-        np.where(wind_t == date_time.floor('hour'))[0])
+    i_date_time = np.where(wind_t == date_time.floor('hour'))[0].item()
     i_date_time_p1 = i_date_time + 1
     u_avg = np.mean(wind_u[(i_date_time_p1 + avg_hrs):i_date_time_p1])
     v_avg = np.mean(wind_v[(i_date_time_p1 + avg_hrs):i_date_time_p1])