ocf-data-sampler 0.5.30__py3-none-any.whl → 0.5.32__py3-none-any.whl

ocf_data_sampler/select/select_spatial_slice.py

@@ -43,104 +43,11 @@ def _get_pixel_index_location(da: xr.DataArray, location: Location) -> tuple[int
     return closest_x, closest_y
 
 
-def _select_padded_slice(
-    da: xr.DataArray,
-    left_idx: int,
-    right_idx: int,
-    bottom_idx: int,
-    top_idx: int,
-    x_dim: str,
-    y_dim: str,
-) -> xr.DataArray:
-    """Selects spatial slice - padding where necessary if indices are out of bounds.
-
-    Args:
-        da: xarray DataArray.
-        left_idx: The leftmost index of the slice.
-        right_idx: The rightmost index of the slice.
-        bottom_idx: The bottommost index of the slice.
-        top_idx: The topmost index of the slice.
-        x_dim: Name of the x dimension.
-        y_dim: Name of the y dimension.
-
-    Returns:
-        An xarray DataArray with padding, if necessary.
-    """
-    data_width_pixels = len(da[x_dim])
-    data_height_pixels = len(da[y_dim])
-
-    left_pad_pixels = max(0, -left_idx)
-    right_pad_pixels = max(0, right_idx - data_width_pixels)
-    bottom_pad_pixels = max(0, -bottom_idx)
-    top_pad_pixels = max(0, top_idx - data_height_pixels)
-
-    if (left_pad_pixels > 0 and right_pad_pixels > 0) or (
-        bottom_pad_pixels > 0 and top_pad_pixels > 0
-    ):
-        raise ValueError("Cannot pad both sides of the window")
-
-    dx = np.median(np.diff(da[x_dim].values))
-    dy = np.median(np.diff(da[y_dim].values))
-
-    # Create a new DataArray which has indices which go outside
-    # the original DataArray
-    # Pad the left of the window
-    if left_pad_pixels > 0:
-        x_sel = np.concatenate(
-            [
-                da[x_dim].values[0] + np.arange(-left_pad_pixels, 0) * dx,
-                da[x_dim].values[0:right_idx],
-            ],
-        )
-        da = da.isel({x_dim: slice(0, right_idx)}).reindex({x_dim: x_sel})
-
-    # Pad the right of the window
-    elif right_pad_pixels > 0:
-        x_sel = np.concatenate(
-            [
-                da[x_dim].values[left_idx:],
-                da[x_dim].values[-1] + np.arange(1, right_pad_pixels + 1) * dx,
-            ],
-        )
-        da = da.isel({x_dim: slice(left_idx, None)}).reindex({x_dim: x_sel})
-
-    # No left-right padding required
-    else:
-        da = da.isel({x_dim: slice(left_idx, right_idx)})
-
-    # Pad the bottom of the window
-    if bottom_pad_pixels > 0:
-        y_sel = np.concatenate(
-            [
-                da[y_dim].values[0] + np.arange(-bottom_pad_pixels, 0) * dy,
-                da[y_dim].values[0:top_idx],
-            ],
-        )
-        da = da.isel({y_dim: slice(0, top_idx)}).reindex({y_dim: y_sel})
-
-    # Pad the top of the window
-    elif top_pad_pixels > 0:
-        y_sel = np.concatenate(
-            [
-                da[y_dim].values[bottom_idx:],
-                da[y_dim].values[-1] + np.arange(1, top_pad_pixels + 1) * dy,
-            ],
-        )
-        da = da.isel({y_dim: slice(bottom_idx, None)}).reindex({y_dim: y_sel})
-
-    # No bottom-top padding required
-    else:
-        da = da.isel({y_dim: slice(bottom_idx, top_idx)})
-
-    return da
-
-
 def select_spatial_slice_pixels(
     da: xr.DataArray,
     location: Location,
     width_pixels: int,
     height_pixels: int,
-    allow_partial_slice: bool = False,
 ) -> xr.DataArray:
     """Select spatial slice based off pixels from location point of interest.
 
@@ -149,7 +56,6 @@ def select_spatial_slice_pixels(
         location: Location of interest that will be the center of the returned slice
         height_pixels: Height of the slice in pixels
         width_pixels: Width of the slice in pixels
-        allow_partial_slice: Whether to allow a partial slice.
 
     Returns:
         The selected DataArray slice.
@@ -157,7 +63,6 @@ def select_spatial_slice_pixels(
     Raises:
         ValueError: If the dimensions are not even or the slice is not allowed
                     when padding is required.
-
     """
     if (width_pixels % 2) != 0:
         raise ValueError("Width must be an even number")
@@ -179,39 +84,27 @@ def select_spatial_slice_pixels(
     data_height_pixels = len(da[y_dim])
 
     # Padding checks
-    pad_required = (
+    slice_unavailable = (
         left_idx < 0
         or right_idx > data_width_pixels
         or bottom_idx < 0
         or top_idx > data_height_pixels
     )
 
-    if pad_required:
-        if allow_partial_slice:
-            da = _select_padded_slice(da, left_idx, right_idx, bottom_idx, top_idx, x_dim, y_dim)
-        else:
-            issues = []
-            if left_idx < 0:
-                issues.append(f"left_idx ({left_idx}) < 0")
-            if right_idx > data_width_pixels:
-                issues.append(f"right_idx ({right_idx}) > data_width_pixels ({data_width_pixels})")
-            if bottom_idx < 0:
-                issues.append(f"bottom_idx ({bottom_idx}) < 0")
-            if top_idx > data_height_pixels:
-                issues.append(f"top_idx ({top_idx}) > data_height_pixels ({data_height_pixels})")
-            issue_details = "\n".join(issues)
-            raise ValueError(
-                f"Window for location {location} not available.  Padding required due to: \n"
-                f"{issue_details}\n"
-                "You may wish to set `allow_partial_slice=True`",
-            )
-    else:
-        # Standard selection - without padding
-        da = da.isel({x_dim: slice(left_idx, right_idx), y_dim: slice(bottom_idx, top_idx)})
-
-    if len(da[x_dim]) != width_pixels:
-        raise ValueError(f"x-dim has size {len(da[x_dim])}, expected {width_pixels}")
-    if len(da[y_dim]) != height_pixels:
-        raise ValueError(f"y-dim has size {len(da[y_dim])}, expected {height_pixels}")
+    if slice_unavailable:
+        issues = []
+        if left_idx < 0:
+            issues.append(f"left_idx ({left_idx}) < 0")
+        if right_idx > data_width_pixels:
+            issues.append(f"right_idx ({right_idx}) > data_width_pixels ({data_width_pixels})")
+        if bottom_idx < 0:
+            issues.append(f"bottom_idx ({bottom_idx}) < 0")
+        if top_idx > data_height_pixels:
+            issues.append(f"top_idx ({top_idx}) > data_height_pixels ({data_height_pixels})")
+        issue_details = "\n - ".join(issues)
+        raise ValueError(f"Window for location {location} not available: \n - {issue_details}")
+
+    # Standard selection - without padding
+    da = da.isel({x_dim: slice(left_idx, right_idx), y_dim: slice(bottom_idx, top_idx)})
 
     return da

ocf_data_sampler/torch_datasets/datasets/site.py

@@ -1,6 +1,5 @@
 """Torch dataset for sites."""
 
-import numpy as np
 import pandas as pd
 import xarray as xr
 from torch.utils.data import Dataset
@@ -436,25 +435,3 @@ class SitesDatasetConcurrent(PickleCacheMixin, Dataset):
             site_samples.append(site_numpy_sample)
 
         return stack_np_samples_into_batch(site_samples)
-
-
-def coarsen_data(xr_data: xr.Dataset, coarsen_to_deg: float = 0.1) -> xr.Dataset:
-    """Coarsen the data to a specified resolution in degrees.
-
-    Args:
-        xr_data: xarray dataset to coarsen
-        coarsen_to_deg: resolution to coarsen to in degrees
-    """
-    if "latitude" in xr_data.coords and "longitude" in xr_data.coords:
-        step = np.abs(xr_data.latitude.values[1] - xr_data.latitude.values[0])
-        step = np.round(step, 4)
-        coarsen_factor = int(coarsen_to_deg / step)
-        if coarsen_factor > 1:
-            xr_data = xr_data.coarsen(
-                latitude=coarsen_factor,
-                longitude=coarsen_factor,
-                boundary="pad",
-                coord_func="min",
-            ).mean()
-
-    return xr_data

{ocf_data_sampler-0.5.30.dist-info → ocf_data_sampler-0.5.32.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ocf-data-sampler
-Version: 0.5.30
+Version: 0.5.32
 Author: James Fulton, Peter Dudfield
 Author-email: Open Climate Fix team <info@openclimatefix.org>
 License: MIT License

{ocf_data_sampler-0.5.30.dist-info → ocf_data_sampler-0.5.32.dist-info}/RECORD

@@ -38,12 +38,12 @@ ocf_data_sampler/select/fill_time_periods.py,sha256=TlGxp1xiAqnhdWfLy0pv3FuZc00d
 ocf_data_sampler/select/find_contiguous_time_periods.py,sha256=etkr6LuB7zxkfzWJ6SgHiULdRuFzFlq5bOUNd257Qx4,11545
 ocf_data_sampler/select/geospatial.py,sha256=rvMy_e--3tm-KAy9pU6b9-UMBQqH2sXykr3N_4SHYy4,6528
 ocf_data_sampler/select/location.py,sha256=nicHRegnD5QPJ3-4C9vH33bIcLowjfT0xwEF6sZNOOY,2348
-ocf_data_sampler/select/select_spatial_slice.py,sha256=NB6NtZBc_Mb5zPCItzBIEa_Nroj2kEsjUIsa_kdWoj0,7081
+ocf_data_sampler/select/select_spatial_slice.py,sha256=Nov6foM5xPkAREsEAHHAak8jHlksC-wAoxcso48D4wQ,3503
 ocf_data_sampler/select/select_time_slice.py,sha256=cpkdovJMvcjxSGfq9G0OJK5aDAeCXg7exWYrJnR4N2w,4116
 ocf_data_sampler/torch_datasets/datasets/__init__.py,sha256=o0SsEXXZ6k9iL__5_RN1Sf60lw_eqK91P3UFEHAD2k0,102
 ocf_data_sampler/torch_datasets/datasets/picklecache.py,sha256=b8T5lgKfiPXLwuVQuFpCQBlU-HNBrA-Z-eSwYICKvsQ,1350
 ocf_data_sampler/torch_datasets/datasets/pvnet_uk.py,sha256=Tpf1zVPtBy-815n__-zgsTeNRxlxjLwuu6UgCCyIEL0,12365
-ocf_data_sampler/torch_datasets/datasets/site.py,sha256=9WOEMkqqzG_SziP3uJ7YFBKnG8Gu7yHolB0UFNLyvEg,15877
+ocf_data_sampler/torch_datasets/datasets/site.py,sha256=F3emPejGQNtm3qzWtIGSFOVVH9l3DxS9YkiSATXROHQ,15095
 ocf_data_sampler/torch_datasets/sample/__init__.py,sha256=GL84vdZl_SjHDGVyh9Uekx2XhPYuZ0dnO3l6f6KXnHI,100
 ocf_data_sampler/torch_datasets/sample/base.py,sha256=cQ1oIyhdmlotejZK8B3Cw6MNvpdnBPD8G_o2h7Ye4Vc,2206
 ocf_data_sampler/torch_datasets/sample/site.py,sha256=40NwNTqjL1WVhPdwe02zDHHfDLG2u_bvCfRCtGAtFc0,1466
@@ -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-0.5.30.dist-info/METADATA,sha256=UrhYQvOnfGclWu1nrAouW0NMH6uWVOqBK7Au62W3jXM,13541
-ocf_data_sampler-0.5.30.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-ocf_data_sampler-0.5.30.dist-info/top_level.txt,sha256=deUxqmsONNAGZDNbsntbXH7BRA1MqWaUeAJrCo6q_xA,25
-ocf_data_sampler-0.5.30.dist-info/RECORD,,
+ocf_data_sampler-0.5.32.dist-info/METADATA,sha256=4RiXFoeJfv5q6IiLzSgJpU3sAVDDjtnxC3iNhWBRv5A,13541
+ocf_data_sampler-0.5.32.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+ocf_data_sampler-0.5.32.dist-info/top_level.txt,sha256=deUxqmsONNAGZDNbsntbXH7BRA1MqWaUeAJrCo6q_xA,25
+ocf_data_sampler-0.5.32.dist-info/RECORD,,