Merge pull request #276 from sentinel-hub/feat/fix-dependecy-updates

Fix dependecy updates

Author: AleksMat

core/eolearn/core/eodata.py

@@ -736,7 +736,7 @@ def _parse_feature_value(self, value):
                                  'dimension{}'.format(self.feature_type, self.ndim, 's' if self.ndim > 1 else ''))
 
             if self.feature_type.is_discrete():
-                if not issubclass(value.dtype.type, (np.integer, np.bool, np.bool_, np.bool8)):
+                if not issubclass(value.dtype.type, (np.integer, bool, np.bool_, np.bool8)):
                     msg = '{} is a discrete feature type therefore dtype of data should be a subtype of ' \
                           'numpy.integer or numpy.bool, found type {}. In the future an error will be raised because ' \
                           'of this'.format(self.feature_type, value.dtype.type)

core/eolearn/tests/test_eodata.py

@@ -37,7 +37,7 @@ def test_numpy_feature_types(self):
 
         data_examples = []
         for size in range(6):
-            for dtype in [np.float32, np.float64, np.float, np.uint8, np.int64, np.bool]:
+            for dtype in [np.float32, np.float64, float, np.uint8, np.int64, bool]:
                 data_examples.append(np.zeros((2, ) * size, dtype=dtype))
 
         for feature_type in FeatureTypeSet.RASTER_TYPES:
@@ -252,8 +252,8 @@ def test_concatenate_missmatched_timeless(self):
             _ = EOPatch.concatenate(eop1, eop2)
 
     def test_equals(self):
-        eop1 = EOPatch(data={'bands': np.arange(2 * 3 * 3 * 2).reshape(2, 3, 3, 2)})
-        eop2 = EOPatch(data={'bands': np.arange(2 * 3 * 3 * 2).reshape(2, 3, 3, 2)})
+        eop1 = EOPatch(data={'bands': np.arange(2 * 3 * 3 * 2, dtype=np.float32).reshape(2, 3, 3, 2)})
+        eop2 = EOPatch(data={'bands': np.arange(2 * 3 * 3 * 2, dtype=np.float32).reshape(2, 3, 3, 2)})
         self.assertEqual(eop1, eop2)
 
         eop1.data['bands'][1, ...] = np.nan

coregistration/eolearn/coregistration/coregistration.py

@@ -141,7 +141,7 @@ def execute(self, eopatch):
             sliced_data = [np.clip(d, np.percentile(d, 5), np.percentile(d, 95)) for d in sliced_data]
         except BaseException:
             LOGGER.warning("Could not calculate gradients, using original data")
-            sliced_data = [d for d in sliced_data]
+            sliced_data = list(sliced_data)
 
         iflag = self._get_interpolation_flag(self.interpolation_type)
 

features/eolearn/features/blob.py

@@ -73,13 +73,13 @@ def __init__(self, feature, blob_object, **blob_parameters):
         self.blob_parameters = blob_parameters
 
     def _compute_blob(self, data):
-        result = np.zeros(data.shape, dtype=np.float)
+        result = np.zeros(data.shape, dtype=float)
         for time in range(data.shape[0]):
             for band in range(data.shape[-1]):
                 image = data[time, :, :, band]
                 res = np.asarray(self.blob_object(image, **self.blob_parameters))
-                x_coord = res[:, 0].astype(np.int)
-                y_coord = res[:, 1].astype(np.int)
+                x_coord = res[:, 0].astype(int)
+                y_coord = res[:, 1].astype(int)
                 radius = res[:, 2] * sqrt(2)
                 result[time, x_coord, y_coord, band] = radius
         return result

features/eolearn/features/haralick.py

@@ -152,7 +152,7 @@ def _custom_texture(self, glcm):
         return res
 
     def _calculate_haralick(self, data):
-        result = np.empty(data.shape, dtype=np.float)
+        result = np.empty(data.shape, dtype=float)
         # For each date and each band
         for time in range(data.shape[0]):
             for band in range(data.shape[3]):

features/eolearn/features/hog.py

@@ -62,7 +62,7 @@ def _compute_hog(self, data):
                                (int(data.shape[1] // self.pixels_per_cell[0]) - self.cells_per_block[0] + 1) *
                                self.cells_per_block[0],
                                (int(data.shape[2] // self.pixels_per_cell[1]) - self.cells_per_block[1] + 1) *
-                               self.cells_per_block[1], self.n_orientations), dtype=np.float)
+                               self.cells_per_block[1], self.n_orientations), dtype=float)
         if self.visualize:
             im_visu = np.empty(data.shape[0:3] + (1,))
         for time in range(data.shape[0]):

features/eolearn/features/interpolation.py

@@ -317,8 +317,8 @@ def interpolate_data(self, data, times, resampled_times):
         # find NaNs that start or end a time-series
         row_nans, col_nans = np.where(self._get_start_end_nans(data))
         nan_row_res_indices = np.array([index for index in ori2res[row_nans] if index is not None], dtype=np.int32)
-        nan_col_res_indices = np.array([index is not None for index in ori2res[row_nans]],
-                                       dtype=np.bool)
+        nan_col_res_indices = np.array([index is not None for index in ori2res[row_nans]], dtype=bool)
+
         if nan_row_res_indices.size:
             # mask out from output values the starting/ending NaNs
             res_temp_values[nan_row_res_indices, col_nans[nan_col_res_indices]] = np.nan
@@ -400,7 +400,7 @@ def execute(self, eopatch):
         # Apply a mask on data
         if self.mask_feature is not None:
             for mask_type, mask_name in self.mask_feature(eopatch):
-                negated_mask = ~eopatch[mask_type][mask_name].astype(np.bool)
+                negated_mask = ~eopatch[mask_type][mask_name].astype(bool)
                 feature_data = self._mask_feature_data(feature_data, negated_mask, mask_type)
 
         # Flatten array

features/eolearn/features/local_binary_pattern.py

@@ -44,7 +44,7 @@ def __init__(self, feature, nb_points=24, radius=3):
             raise ValueError('Local binary pattern task parameters must be positives')
 
     def _compute_lbp(self, data):
-        result = np.empty(data.shape, dtype=np.float)
+        result = np.empty(data.shape, dtype=float)
         for time in range(data.shape[0]):
             for band in range(data.shape[-1]):
                 image = data[time, :, :, band]

features/eolearn/features/radiometric_normalization.py

@@ -122,7 +122,7 @@ def _geoville_index_by_percentile(self, data, percentile):
         k_arr = (valid_obs - 1) * (percentile / 100.0)
         k_arr = np.where(k_arr < 0, 0, k_arr)
         f_arr = np.floor(k_arr + 0.5)
-        f_arr = f_arr.astype(np.int)
+        f_arr = f_arr.astype(int)
         # get floor value of reference band and index band
         ind = f_arr.astype("int16")
         y_val, x_val = ind_tmp.shape[1], ind_tmp.shape[2]

features/eolearn/features/temporal_features.py

@@ -145,7 +145,7 @@ def execute(self, eopatch):
 
         madata = np.ma.array(data,
                              dtype=np.float32,
-                             mask=np.logical_or(~valid_data_mask.astype(np.bool), np.isnan(data)))
+                             mask=np.logical_or(~valid_data_mask.astype(bool), np.isnan(data)))
 
         argmax_data = np.ma.MaskedArray.argmax(madata, axis=0)
         argmin_data = np.ma.MaskedArray.argmin(madata, axis=0)
@@ -206,7 +206,7 @@ def execute(self, eopatch):
 
         ndvi = np.ma.array(eopatch.data[self.data_feature],
                            dtype=np.float32,
-                           mask=~valid_data_mask.astype(np.bool))
+                           mask=~valid_data_mask.astype(bool))
 
         all_dates = np.asarray([x.toordinal() for x in eopatch.timestamp])