[Bug]: Debug zppy diffs for v3.0.0 (#931)

* Fix RMSE and CORR text position in plots
* Fix ref plot max values zonal/meridional mean
* Fix rmse/corr values for zonal_mean_2d
* Fix metrics and fonts in polar
* Fix aerosol_aeronet X and Y axes labels
  - Add `aerosol_aeronet` to `SETS_USING_LAT_LON_FORMATTER`
  - Remove unused `plt.legend()` in `aerosol_aeronet_plot.py` to avoid UserWarning: No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.
* Update `_hybrid_to_plevs()` to keep more attrs
  - Keep `long_name`. `long_name` is used for downstream plotting, including zonal_mean_xy.
  - Update `zonal_mean_xy_plot.py` to use correct `long_name` attribute
* Fix relative diff plots displaying mean nans
  - This was caused by performing arithmetic on  the entire regridded and spatially averaged diffs dataset, which also affected other variables such as bounds
* Add remote attachment config to workspace
---------

Co-authored-by: chengzhuzhang <[email protected]>

Author: tomvothecoder

.vscode/e3sm_diags.code-workspace

@@ -54,6 +54,17 @@
   "launch": {
     "version": "0.2.0",
     "configurations": [
+      {
+        "name": "Python: Current File",
+        "type": "debugpy",
+        "request": "launch",
+        "program": "${file}",
+        "console": "integratedTerminal",
+        "justMyCode": true,
+        "env": {
+          "PYTHONPATH": "${workspaceFolder}"
+        }
+      },
       {
         "name": "Python: Current File",
         "type": "debugpy",

auxiliary_tools/cdat_regression_testing/930-zppy-diffs/v2/debug-large-diffs/debug.py

@@ -0,0 +1,155 @@
+#%%
+import glob
+from typing import List
+
+import numpy as np
+import xarray as xr
+
+from e3sm_diags.derivations.derivations import DERIVED_VARIABLES
+import pandas as pd
+
+DEV_DIR = "25-02-04-branch-930-zppy-diffs"
+DEV_PATH = f"/lcrc/group/e3sm/public_html/cdat-migration-fy24/{DEV_DIR}/"
+
+DEV_GLOB_ALL = sorted(glob.glob(DEV_PATH + "**/**/*.nc"))
+DEV_NUM_FILES = len(DEV_GLOB_ALL)
+
+MAIN_DIR = "25-02-04-main-zppy-diffs"
+MAIN_PATH = f"/lcrc/group/e3sm/public_html/cdat-migration-fy24/{MAIN_DIR}/"
+MAIN_GLOB_ALL = sorted(glob.glob(MAIN_PATH + "**/**/*.nc"))
+MAIN_NUM_FILES = len(MAIN_GLOB_ALL)
+
+#%%
+KEEP_VARS = ["OMI-MLS-TCO-ANN-60S60N"]
+
+DEV_GLOB = [fp for fp in DEV_GLOB_ALL if "diff.nc" not in fp and any(var in fp for var in KEEP_VARS)]
+MAIN_GLOB = [fp for fp in MAIN_GLOB_ALL if "diff.nc" not in fp and any(var in fp for var in KEEP_VARS)]
+
+DEV_GLOB_DIFF = [fp for fp in DEV_GLOB_ALL if "diff.nc" in fp and any(var in fp for var in KEEP_VARS)]
+MAIN_GLOB_DIFF = [fp for fp in MAIN_GLOB_ALL if "diff.nc" in fp and any(var in fp for var in KEEP_VARS)]
+
+def _get_var_data(ds: xr.Dataset, var_key: str) -> np.ndarray:
+    """Get the variable data using a list of matching keys.
+
+    The `main` branch saves the dataset using the original variable name,
+    while the dev branch saves the variable with the derived variable name.
+    The dev branch is performing the expected behavior here.
+
+    Parameters
+    ----------
+    ds : xr.Dataset
+        _description_
+    var_key : str
+        _description_
+
+    Returns
+    -------
+    np.ndarray
+        _description_
+    """
+
+    data = None
+
+    try:
+        data = ds[var_key]
+    except KeyError:
+        try:
+            var_keys = DERIVED_VARIABLES[var_key].keys()
+        except KeyError:
+            var_keys = DERIVED_VARIABLES[var_key.upper()].keys()
+
+        var_keys = [var_key] + list(sum(var_keys, ()))
+
+        for key in var_keys:
+            if key in ds.data_vars.keys():
+                data = ds[key]
+                break
+
+    return data
+#%%
+ATOL = 0
+RTOL = 1e-5
+
+print(f"Relative tolerance: {RTOL}, Absolute tolerance: {ATOL}")
+
+def compare_files(main_glob: List[str]):
+    for fp_main in main_glob:
+        var_key = fp_main.split("-")[-3]
+        fp_type = fp_main.split("-")[-1].split("_")[-1]
+
+        fp_dev = fp_main.replace(MAIN_DIR, DEV_DIR)
+
+        print(f"{var_key} - {fp_type}")
+        print("-" * 50)
+        print(f"Main: {fp_main}\nDev: {fp_dev}")
+
+        ds_main = xr.open_dataset(fp_main)
+        ds_dev = xr.open_dataset(fp_dev)
+
+        dv_main = _get_var_data(ds_main, var_key)
+        dv_dev = _get_var_data(ds_dev, var_key)
+
+        if dv_main is None:
+            dv_main = _get_var_data(ds_main, var_key + "_diff")
+
+        try:
+            np.testing.assert_allclose(dv_main.values, dv_dev.values, rtol=RTOL, atol=ATOL)
+        except AssertionError as e:
+            print(f"{e}\n")
+        else:
+            print(f"Arrays are within relative tolerance.\n")
+
+print("Comparing test.nc and ref.nc files")
+print("=" * 50)
+compare_files(MAIN_GLOB)
+
+#%%
+print("Comparing diff.nc files")
+print("=" * 50)
+compare_files(MAIN_GLOB_DIFF)
+
+# %%
+def compare_stats(main_glob_diff: List[str]):
+    for fp_main in main_glob_diff:
+        var_key = fp_main.split("-")[-3]
+        fp_type = fp_main.split("-")[-1].split("_")[-1]
+
+        fp_dev = fp_main.replace(MAIN_DIR, DEV_DIR)
+
+        print(f"{var_key} - {fp_type}")
+        print("-" * 50)
+        print(f"Main: {fp_main}\nDev: {fp_dev}")
+
+        ds_main = xr.open_dataset(fp_main)
+        ds_dev = xr.open_dataset(fp_dev)
+
+        dv_main = _get_var_data(ds_main, var_key)
+        dv_dev = _get_var_data(ds_dev, var_key)
+
+        if dv_main is None:
+            dv_main = _get_var_data(ds_main, var_key + "_diff")
+            stats_main = {
+                "min": dv_main.min().item(),
+                "max": dv_main.max().item(),
+                "mean": dv_main.mean().item(),
+                "sum": dv_main.sum().item(),
+                "nan_count": np.isnan(dv_main).sum().item()
+            }
+
+            stats_dev = {
+                "min": dv_dev.min().item(),
+                "max": dv_dev.max().item(),
+                "mean": dv_dev.mean().item(),
+                "sum": dv_dev.sum().item(),
+                "nan_count": np.isnan(dv_dev).sum().item()
+            }
+
+            df_stats = pd.DataFrame([stats_main, stats_dev], index=["main", "dev"])
+            print(df_stats)
+
+print("Comparing stats of diff.nc files")
+print("=" * 50)
+compare_stats(MAIN_GLOB_DIFF)
+
+
+# %%