Skip to content

Commit 0466561

Browse files
committed
update docstring
1 parent 1b03d4b commit 0466561

File tree

1 file changed

+2
-2
lines changed

1 file changed

+2
-2
lines changed

stackstac/stack.py

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -168,7 +168,7 @@ def stack(
168168
such as ``1024``, ``(1024, 2048)``, ``(10, "auto", 512, 512)``, ``15 MB``, etc.
169169
170170
If only 1 or 2 sizes are given, like ``2048`` or ``(512, 1024)``, this is used to chunk
171-
just the spatial dimensions. The time and band dimensions will have a chunksize of 1,
171+
just the spatial dimensions (last two). The time and band dimensions will have a chunksize of 1,
172172
meaning that every STAC Asset will be its own chunk. (This is the default.)
173173
174174
If you'll be filtering items somewhat randomly (like ``stack[stack["eo:cloud_cover"] < 20]``),
@@ -180,7 +180,7 @@ def stack(
180180
for the time or band dimensions can help a lot. For example, ``chunksize=(10, 1, 512, 512)`` would
181181
process in 512x512 pixel tiles, loading 10 assets at a time per tile. ``chunksize=(-1, 1, 512, 512)``
182182
would load *every* item within the 512x512 tile into the chunk.
183-
If you'll be immediately compositing the data (like ``.mean("time")``), doing this is
183+
If you'll be immediately compositing the data (like ``.median("time")``), doing this is
184184
often a good idea because you'll be flattening the assets together anyway.
185185
dtype:
186186
The NumPy data type of the output array. Default: ``float64``. Must be a data type

0 commit comments

Comments
 (0)