Issue/233 data validation

nowcasting_dataset/data_sources/fake.py

@@ -241,6 +241,8 @@ def create_gsp_pv_dataset(
     data["x_coords"] = x_coords
     data["y_coords"] = y_coords
 
+    data.__setitem__("data", data.data.clip(min=0))
+
     return data
 
 
@@ -275,6 +277,9 @@ def create_sun_dataset(
     sun = data.rename({"data": "elevation"})
     sun["azimuth"] = data.data
 
+    sun.__setitem__("azimuth", sun.azimuth.clip(min=0, max=360))
+    sun.__setitem__("elevation", sun.elevation.clip(min=-90, max=90))
+
     return sun
 
 

nowcasting_dataset/data_sources/gsp/gsp_data_source.py

@@ -234,9 +234,9 @@ def get_example(
         gsp["x_coords"] = gsp_x_coords
         gsp["y_coords"] = gsp_y_coords
 
-        # pad out so that there are always 32 gsp
+        # pad out so that there are always 32 gsp, fill with 0
         pad_n = self.n_gsp_per_example - len(gsp.id_index)
-        gsp = gsp.pad(id_index=(0, pad_n), data=((0, 0), (0, pad_n)))
+        gsp = gsp.pad(id_index=(0, pad_n), data=((0, 0), (0, pad_n)), constant_values=0)
 
         gsp.__setitem__("id_index", range(self.n_gsp_per_example))
 

nowcasting_dataset/data_sources/gsp/gsp_model.py

@@ -1,5 +1,6 @@
 """ Model for output of GSP data """
 import logging
+import numpy as np
 
 from nowcasting_dataset.data_sources.datasource_output import (
     DataSourceOutput,
@@ -15,4 +16,11 @@ class GSP(DataSourceOutput):
     __slots__ = ()
     _expected_dimensions = ("time", "id")
 
-    # todo add validation here - https://github.com/openclimatefix/nowcasting_dataset/issues/233
+    @classmethod
+    def model_validation(cls, v):
+        """ Check that all values are non NaNs """
+        assert (~np.isnan(v.data)).all(), f"Some gsp data values are NaNs"
+        assert (v.data != np.Inf).all(), f"Some gsp data values are Infinite"
+        assert (v.data >= 0).all(), f"Some gsp data values are below 0 {v.data.min()}"
+
+        return v

nowcasting_dataset/data_sources/nwp/nwp_model.py

@@ -24,5 +24,5 @@ class NWP(DataSourceOutput):
     @classmethod
     def model_validation(cls, v):
         """ Check that all values are not NaNs """
-        assert (v.data != np.nan).all(), "Some nwp data values are NaNs"
+        assert (~np.isnan(v.data)).all(), "Some nwp data values are NaNs"
         return v

nowcasting_dataset/data_sources/pv/pv_data_source.py

@@ -268,9 +268,9 @@ def get_example(
         pv["x_coords"] = x_coords
         pv["y_coords"] = y_coords
 
-        # pad out so that there are always 32 gsp
+        # pad out so that there are always 32 gsp, pad with zeros
         pad_n = self.n_pv_systems_per_example - len(pv.id_index)
-        pv = pv.pad(id_index=(0, pad_n), data=((0, 0), (0, pad_n)))
+        pv = pv.pad(id_index=(0, pad_n), data=((0, 0), (0, pad_n)), constant_values=0)
 
         pv.__setitem__("id_index", range(self.n_pv_systems_per_example))
 

nowcasting_dataset/data_sources/pv/pv_model.py

@@ -29,4 +29,12 @@ class PV(DataSourceOutput):
     __slots__ = ()
     _expected_dimensions = ("time", "id")
 
-    # todo add validation here - https://github.com/openclimatefix/nowcasting_dataset/issues/233
+    @classmethod
+    def model_validation(cls, v):
+        """ Check that all values are non NaNs """
+        assert (~np.isnan(v.data)).all(), f"Some pv data values are NaNs"
+        assert (v.data != np.Inf).all(), f"Some pv data values are Infinite"
+
+        assert (v.data >= 0).all(), f"Some pv data values are below 0"
+
+        return v

nowcasting_dataset/data_sources/satellite/satellite_model.py

@@ -27,5 +27,6 @@ class Satellite(DataSourceOutput):
     @classmethod
     def model_validation(cls, v):
         """ Check that all values are non negative """
-        assert (v.data != np.NaN).all(), f"Some satellite data values are NaNs"
+        assert (~np.isnan(v.data)).all(), f"Some satellite data values are NaNs"
+        assert (v.data != -1).all(), f"Some satellite data values are -1's"
         return v

nowcasting_dataset/data_sources/sun/sun_model.py

@@ -20,4 +20,25 @@ class Sun(DataSourceOutput):
     __slots__ = ()
     _expected_dimensions = ("time",)
 
-    # todo add validation here - https://github.com/openclimatefix/nowcasting_dataset/issues/233
+    @classmethod
+    def model_validation(cls, v):
+        """ Check that all values are non NaNs """
+        assert (~np.isnan(v.elevation)).all(), f"Some elevation data values are NaNs"
+        assert (v.elevation != np.Inf).all(), f"Some elevation data values are Infinite"
+
+        assert (~np.isnan(v.azimuth)).all(), f"Some azimuth data values are NaNs"
+        assert (v.azimuth != np.Inf).all(), f"Some azimuth data values are Infinite"
+
+        assert (0 <= v.azimuth).all(), f"Some azimuth data values are lower 0, {v.azimuth.min()}"
+        assert (
+            v.azimuth <= 360
+        ).all(), f"Some azimuth data values are greater than 360, {v.azimuth.max()}"
+
+        assert (
+            -90 <= v.elevation
+        ).all(), f"Some elevation data values are lower -90, {v.elevation.min()}"
+        assert (
+            v.elevation <= 90
+        ).all(), f"Some elevation data values are greater than 90, {v.elevation.max()}"
+
+        return v

nowcasting_dataset/data_sources/topographic/topographic_model.py

@@ -21,5 +21,6 @@ class Topographic(DataSourceOutput):
     @classmethod
     def model_validation(cls, v):
         """ Check that all values are non NaNs """
-        assert (v.data != np.NaN).all(), f"Some topological data values are NaNs"
+        assert (~np.isnan(v.data)).all(), f"Some topological data values are NaNs"
+        assert (v.data != np.Inf).all(), f"Some topological data values are Infinite"
         return v

tests/data_sources/gsp/test_gsp_model.py

@@ -0,0 +1,30 @@
+import os
+import tempfile
+import pytest
+import numpy as np
+
+from nowcasting_dataset.data_sources.fake import gsp_fake
+from nowcasting_dataset.data_sources.gsp.gsp_model import GSP
+
+
+def test_gsp_init():
+    _ = gsp_fake(batch_size=4, seq_length_30=5, n_gsp_per_batch=6)
+
+
+def test_gsp_validation():
+    gsp = gsp_fake(batch_size=4, seq_length_30=5, n_gsp_per_batch=6)
+
+    GSP.model_validation(gsp)
+
+    gsp.data[0, 0] = np.nan
+    with pytest.raises(Exception):
+        GSP.model_validation(gsp)
+
+
+def test_gsp_save():
+
+    with tempfile.TemporaryDirectory() as dirpath:
+        gsp = gsp_fake(batch_size=4, seq_length_30=5, n_gsp_per_batch=6)
+        gsp.save_netcdf(path=dirpath, batch_i=0)
+
+        assert os.path.exists(f"{dirpath}/gsp/0.nc")

tests/data_sources/satellite/test_satellite_model.py

@@ -1,11 +1,24 @@
 import os
 import tempfile
+import pytest
+import numpy as np
 
 from nowcasting_dataset.data_sources.fake import satellite_fake
+from nowcasting_dataset.data_sources.satellite.satellite_model import Satellite
 
 
 def test_satellite_init():
-    _ = satellite_fake
+    _ = satellite_fake()
+
+
+def test_satellite_validation():
+    sat = satellite_fake()
+
+    Satellite.model_validation(sat)
+
+    sat.data[0, 0] = np.nan
+    with pytest.raises(Exception):
+        Satellite.model_validation(sat)
 
 
 def test_satellite_save():

tests/data_sources/sun/test_sun_data_source.py

@@ -1,8 +1,4 @@
 from nowcasting_dataset.data_sources.sun.sun_data_source import SunDataSource
-from datetime import datetime
-
-# from nowcasting_dataset.dataset.example import Example
-from nowcasting_dataset.consts import SUN_ELEVATION_ANGLE, SUN_AZIMUTH_ANGLE
 import pandas as pd
 
 

tests/data_sources/sun/test_sun_model.py

@@ -0,0 +1,50 @@
+import os
+import tempfile
+import pytest
+import numpy as np
+
+from nowcasting_dataset.data_sources.fake import sun_fake
+from nowcasting_dataset.data_sources.sun.sun_model import Sun
+
+
+def test_sun_init():
+    _ = sun_fake(batch_size=4, seq_length_5=17)
+
+
+def test_sun_validation():
+    sun = sun_fake(batch_size=4, seq_length_5=17)
+
+    Sun.model_validation(sun)
+
+    sun.elevation[0, 0] = np.nan
+    with pytest.raises(Exception):
+        Sun.model_validation(sun)
+
+
+def test_sun_validation_elevation():
+    sun = sun_fake(batch_size=4, seq_length_5=17)
+
+    Sun.model_validation(sun)
+
+    sun.elevation[0, 0] = 1000
+    with pytest.raises(Exception):
+        Sun.model_validation(sun)
+
+
+def test_sun_validation_azimuth():
+    sun = sun_fake(batch_size=4, seq_length_5=17)
+
+    Sun.model_validation(sun)
+
+    sun.azimuth[0, 0] = 1000
+    with pytest.raises(Exception):
+        Sun.model_validation(sun)
+
+
+def test_sun_save():
+
+    with tempfile.TemporaryDirectory() as dirpath:
+        sun = sun_fake(batch_size=4, seq_length_5=17)
+        sun.save_netcdf(path=dirpath, batch_i=0)
+
+        assert os.path.exists(f"{dirpath}/sun/0.nc")