ocf-data-sampler 0.5.25__py3-none-any.whl → 0.5.26__py3-none-any.whl

ocf_data_sampler/load/site.py

@@ -31,9 +31,10 @@ def open_site(generation_file_path: str, metadata_file_path: str) -> xr.DataArra
         capacity_kwp=("site_id", metadata_df["capacity_kwp"].values),
     )
 
-    # Sanity checks
-    if not np.isfinite(generation_ds.generation_kw.values).all():
-        raise ValueError("generation_kw contains non-finite values")
+    # Sanity checks, to prevent inf or negative values
+    # Note NaNs are allowed in generation_kw as can have non overlapping time periods for sites
+    if np.isinf(generation_ds.generation_kw.values).all():
+        raise ValueError("generation_kw contains infinite (+/- inf) values")
     if not (generation_ds.capacity_kwp.values > 0).all():
         raise ValueError("capacity_kwp contains non-positive values")
 
@@ -43,17 +44,21 @@ def open_site(generation_file_path: str, metadata_file_path: str) -> xr.DataArra
     if not np.issubdtype(site_da.dtype, np.floating):
         raise TypeError(f"Generation data should be float, not {site_da.dtype}")
 
+
     coord_dtypes = {
-        "time_utc": np.datetime64,
-        "site_id": np.integer,
-        "capacity_kwp": np.floating,
-        "latitude": np.floating,
-        "longitude": np.floating,
-    }
-
-    for coord, expected_dtype in coord_dtypes.items():
-        if not np.issubdtype(site_da.coords[coord].dtype, expected_dtype):
+    "time_utc": (np.datetime64,),
+    "site_id": (np.integer,),
+    "capacity_kwp": (np.integer, np.floating),
+    "latitude": (np.floating,),
+    "longitude": (np.floating,),
+}
+    for coord, expected_dtypes in coord_dtypes.items():
+        if not any(np.issubdtype(site_da.coords[coord].dtype, dt) for dt in expected_dtypes):
             dtype = site_da.coords[coord].dtype
-            raise TypeError(f"{coord} should be {expected_dtype.__name__}, not {dtype}")
+            allowed = ", ".join(dt.__name__ for dt in expected_dtypes)
+            raise TypeError(f"{coord} should be one of ({allowed}), not {dtype}")
 
-    return site_da
+    # Load the data eagerly into memory by calling compute
+    # this makes the dataset faster to sample from, but
+    # at the cost of a little extra memory usage
+    return site_da.compute()

{ocf_data_sampler-0.5.25.dist-info → ocf_data_sampler-0.5.26.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ocf-data-sampler
-Version: 0.5.25
+Version: 0.5.26
 Author: James Fulton, Peter Dudfield
 Author-email: Open Climate Fix team <info@openclimatefix.org>
 License: MIT License

{ocf_data_sampler-0.5.25.dist-info → ocf_data_sampler-0.5.26.dist-info}/RECORD

@@ -11,7 +11,7 @@ ocf_data_sampler/load/gsp.py,sha256=zsQ39dZBS45qd86lGfCZUjheLRTtMzIUozj-j8c87UQ,
 ocf_data_sampler/load/load_dataset.py,sha256=K8rWykjII-3g127If7WRRFivzHNx3SshCvZj4uQlf28,2089
 ocf_data_sampler/load/open_xarray_tensorstore.py,sha256=YglCBeKa4mSjUU5qlcMOLZXUtFrPFWVKDeKHLjs_YbA,6353
 ocf_data_sampler/load/satellite.py,sha256=5o5SfcplQfZFlm3JJq73j8_m_cWKpFtKk0tTKGjjCuE,1856
-ocf_data_sampler/load/site.py,sha256=WtOy20VMHJIY0IwEemCdcecSDUGcVaLUown-4ixJw90,2147
+ocf_data_sampler/load/site.py,sha256=bpFABjpvlstn6yJ6OPVlPZms-CjJdxNwCkQafpnj0Ik,2539
 ocf_data_sampler/load/utils.py,sha256=AGL0aOOQPrgqNBTjlBtR7Qg1PyQov3DFJo-y198u8pY,2044
 ocf_data_sampler/load/nwp/__init__.py,sha256=SmcrnbygO5xtCKmGR4wtHrj-HI7nOAvnAtfuvRufBGQ,25
 ocf_data_sampler/load/nwp/nwp.py,sha256=0E9shei3Mq1N7F-fBlEKY5Hm0_kI7ysY_rffnWIshvk,3612
@@ -59,7 +59,7 @@ ocf_data_sampler/torch_datasets/utils/valid_time_periods.py,sha256=xcy75cVxl0Wrg
 ocf_data_sampler/torch_datasets/utils/validation_utils.py,sha256=YqmT-lExWlI8_ul3l0EP73Ik002fStr_bhsZh9mQqEU,4735
 scripts/download_gsp_location_data.py,sha256=rRDXMoqX-RYY4jPdxhdlxJGhWdl6r245F5UARgKV6P4,3121
 scripts/refactor_site.py,sha256=skzvsPP0Cn9yTKndzkilyNcGz4DZ88ctvCJ0XrBdc2A,3135
-ocf_data_sampler-0.5.25.dist-info/METADATA,sha256=9JPhGShknN0Rr_r09f-UqymH1i_DIssLHR2Q8N1bfSE,13541
-ocf_data_sampler-0.5.25.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-ocf_data_sampler-0.5.25.dist-info/top_level.txt,sha256=deUxqmsONNAGZDNbsntbXH7BRA1MqWaUeAJrCo6q_xA,25
-ocf_data_sampler-0.5.25.dist-info/RECORD,,
+ocf_data_sampler-0.5.26.dist-info/METADATA,sha256=NQUuT_DW3MI5F3ZFyupsHs3nfTLLU2Fo2PYzn9PxNqM,13541
+ocf_data_sampler-0.5.26.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+ocf_data_sampler-0.5.26.dist-info/top_level.txt,sha256=deUxqmsONNAGZDNbsntbXH7BRA1MqWaUeAJrCo6q_xA,25
+ocf_data_sampler-0.5.26.dist-info/RECORD,,