Merge pull request #127 from RTIInternational/126-incorrect-weighted-…

…average-calculation-in-compute_zonal_mean 126 incorrect weighted average calculation in compute zonal mean
RTIInternational · Mar 12, 2024 · afbadc8 · afbadc8
2 parents 3d96e9a + 714256c
commit afbadc8
Show file tree

Hide file tree

Showing 19 changed files with 115 additions and 267 deletions.
diff --git a/README.md b/README.md
@@ -42,8 +42,8 @@ $ poetry add git+https://github.com/RTIInternational/teehr.git#[BRANCH TAG]
 
 Use Docker
 ```bash
-$ docker build -t teehr:v0.3.9 .
-$ docker run -it --rm --volume $HOME:$HOME -p 8888:8888 teehr:v0.3.9 jupyter lab --ip 0.0.0.0 $HOME
+$ docker build -t teehr:v0.3.10 .
+$ docker run -it --rm --volume $HOME:$HOME -p 8888:8888 teehr:v0.3.10 jupyter lab --ip 0.0.0.0 $HOME
 ```
 
 ## Examples

diff --git a/docs/sphinx/changelog/index.rst b/docs/sphinx/changelog/index.rst
@@ -1,6 +1,18 @@
 Release Notes
 =============
 
+0.3.10 - 2024-03-07
+--------------------
+
+Added
+^^^^^
+* Added `test_zonal_mean_results.py`
+
+Changed
+^^^^^^^
+* Fixed the calculation of the zonal mean of pixel values in `compute_zonal_mean()` so it caculates
+  the weighted average (divides by the sum of weight values).
+* Updated grid loading tests and data to reflect the fixed method.
 
 0.3.9 - 2024-02-15
 --------------------

diff --git a/poetry.lock b/poetry.lock
diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "teehr"
-version = "0.3.9"
+version = "0.3.10"
 description = "Tools for Exploratory Evaluation in Hydrologic Research"
 authors = [
     "RTI International",
@@ -52,6 +52,7 @@ flake8-docstrings = "^1.7.0"
 nbsphinx = "^0.9.3"
 pandoc = "^2.3"
 myst-nb = "^1.0.0"
+nbstripout = "^0.7.1"
 
 [tool.numpydoc_validation]
 checks = [

diff --git a/src/teehr/loading/grid_loading_example.ipynb b/src/teehr/loading/grid_loading_example.ipynb
diff --git a/src/teehr/loading/nwm/grid_utils.py b/src/teehr/loading/nwm/grid_utils.py
@@ -14,7 +14,7 @@
 def compute_zonal_mean(
     da: xr.DataArray, weights_filepath: str
 ) -> pd.DataFrame:
-    """Compute zonal mean of area-weighted pixels for given
+    """Compute zonal mean (weighted average) of area-weighted pixels for given
     zones and weights."""
     # Read weights file
     weights_df = pd.read_parquet(
@@ -28,12 +28,14 @@ def compute_zonal_mean(
     cols = weights_df.col.values
     # Get the values and apply weights
     var_values = arr_2d[rows, cols]
-    weights_df["value"] = var_values * weights_df.weight.values
-    # Compute mean
-    df = weights_df.groupby(by="location_id")["value"].mean().to_frame()
-    df.reset_index(inplace=True)
+    weights_df["weighted_value"] = var_values * weights_df.weight.values
 
-    return df
+    # Compute weighted average
+    df = weights_df.groupby(
+        by="location_id", as_index=False)[["weighted_value", "weight"]].sum()
+    df["value"] = df.weighted_value/df.weight
+
+    return df[["location_id", "value"]]
 
 
 @dask.delayed