ocf-data-sampler 0.2.2__tar.gz → 0.2.4__tar.gz

{ocf_data_sampler-0.2.2 → ocf_data_sampler-0.2.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: ocf-data-sampler
-Version: 0.2.2
+Version: 0.2.4
 Author: James Fulton, Peter Dudfield
 Author-email: Open Climate Fix team <info@openclimatefix.org>
 License: MIT License

{ocf_data_sampler-0.2.2 → ocf_data_sampler-0.2.4}/ocf_data_sampler/config/model.py

@@ -14,6 +14,7 @@ NWP_PROVIDERS = [
     "ecmwf",
     "gfs",
     "icon_eu",
+    "cloudcasting",
 ]
 
 

{ocf_data_sampler-0.2.2 → ocf_data_sampler-0.2.4}/ocf_data_sampler/load/nwp/nwp.py

@@ -2,6 +2,7 @@
 
 import xarray as xr
 
+from ocf_data_sampler.load.nwp.providers.cloudcasting import open_cloudcasting
 from ocf_data_sampler.load.nwp.providers.ecmwf import open_ifs
 from ocf_data_sampler.load.nwp.providers.gfs import open_gfs
 from ocf_data_sampler.load.nwp.providers.icon import open_icon_eu
@@ -28,6 +29,8 @@ def open_nwp(zarr_path: str | list[str], provider: str) -> xr.DataArray:
         _open_nwp = open_icon_eu
     elif provider == "gfs":
         _open_nwp = open_gfs
+    elif provider == "cloudcasting":
+        _open_nwp = open_cloudcasting
     else:
         raise ValueError(f"Unknown provider: {provider}")
 

ocf_data_sampler-0.2.4/ocf_data_sampler/load/nwp/providers/cloudcasting.py

@@ -0,0 +1,51 @@
+"""Cloudcasting provider loader."""
+
+from pathlib import Path
+
+import xarray as xr
+
+from ocf_data_sampler.load.nwp.providers.utils import open_zarr_paths
+from ocf_data_sampler.load.utils import (
+    check_time_unique_increasing,
+    get_xr_data_array_from_xr_dataset,
+    make_spatial_coords_increasing,
+)
+
+
+def open_cloudcasting(zarr_path: Path | str | list[Path] | list[str]) -> xr.DataArray:
+    """Opens the satellite predictions from cloudcasting.
+
+    Cloudcasting is a OCF forecast product. We forecast future satellite images from recent
+    satellite images. More information can be found in the references below.
+
+    Args:
+        zarr_path: Path to the zarr to open
+
+    Returns:
+        Xarray DataArray of the cloudcasting data
+
+    References:
+            [1] https://www.openclimatefix.org/projects/cloud-forecasting
+            [2] https://github.com/ClimeTrend/cloudcasting
+            [3] https://github.com/openclimatefix/sat_pred
+    """
+    # Open the data
+    ds = open_zarr_paths(zarr_path)
+
+    # Rename
+    ds = ds.rename(
+        {
+            "init_time": "init_time_utc",
+            "variable": "channel",
+        },
+    )
+
+    # Check the timestamps are unique and increasing
+    check_time_unique_increasing(ds.init_time_utc)
+
+    # Make sure the spatial coords are in increasing order
+    ds = make_spatial_coords_increasing(ds, x_coord="x_geostationary", y_coord="y_geostationary")
+
+    ds = ds.transpose("init_time_utc", "step", "channel", "x_geostationary", "y_geostationary")
+
+    return get_xr_data_array_from_xr_dataset(ds)

{ocf_data_sampler-0.2.2 → ocf_data_sampler-0.2.4}/ocf_data_sampler/select/select_time_slice.py

@@ -55,7 +55,10 @@ def select_time_slice_nwp(
     if accum_channels is None:
         accum_channels = []
 
-    if dropout_timedeltas is not None:
+    if dropout_timedeltas is None:
+        dropout_timedeltas = []
+
+    if len(dropout_timedeltas)>0:
         if not all(t < pd.Timedelta(0) for t in dropout_timedeltas):
             raise ValueError("dropout timedeltas must be negative")
         if len(dropout_timedeltas) < 1:
@@ -64,7 +67,7 @@ def select_time_slice_nwp(
     if not (0 <= dropout_frac <= 1):
         raise ValueError("dropout_frac must be between 0 and 1")
 
-    consider_dropout = (dropout_timedeltas is not None) and dropout_frac > 0
+    consider_dropout = len(dropout_timedeltas) > 0 and dropout_frac > 0
 
     # The accumatated and non-accumulated channels
     accum_channels = np.intersect1d(da.channel.values, accum_channels)

{ocf_data_sampler-0.2.2 → ocf_data_sampler-0.2.4}/ocf_data_sampler/torch_datasets/utils/valid_time_periods.py

@@ -28,7 +28,7 @@ def find_valid_time_periods(datasets_dict: dict, config: Configuration) -> pd.Da
         for nwp_key, nwp_config in config.input_data.nwp.items():
             da = datasets_dict["nwp"][nwp_key]
 
-            if nwp_config.dropout_timedeltas_minutes is None:
+            if nwp_config.dropout_timedeltas_minutes==[]:
                 max_dropout = minutes(0)
             else:
                 max_dropout = minutes(np.max(np.abs(nwp_config.dropout_timedeltas_minutes)))

{ocf_data_sampler-0.2.2 → ocf_data_sampler-0.2.4}/ocf_data_sampler.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: ocf-data-sampler
-Version: 0.2.2
+Version: 0.2.4
 Author: James Fulton, Peter Dudfield
 Author-email: Open Climate Fix team <info@openclimatefix.org>
 License: MIT License

{ocf_data_sampler-0.2.2 → ocf_data_sampler-0.2.4}/ocf_data_sampler.egg-info/SOURCES.txt

@@ -22,6 +22,7 @@ ocf_data_sampler/load/utils.py
 ocf_data_sampler/load/nwp/__init__.py
 ocf_data_sampler/load/nwp/nwp.py
 ocf_data_sampler/load/nwp/providers/__init__.py
+ocf_data_sampler/load/nwp/providers/cloudcasting.py
 ocf_data_sampler/load/nwp/providers/ecmwf.py
 ocf_data_sampler/load/nwp/providers/gfs.py
 ocf_data_sampler/load/nwp/providers/icon.py