discontinuum 1.0.0__tar.gz → 1.0.1__tar.gz

{discontinuum-1.0.0 → discontinuum-1.0.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: discontinuum
-Version: 1.0.0
+Version: 1.0.1
 Summary: Estimate discontinuous timeseries from continuous covariates.
 Maintainer-email: Timothy Hodson <thodson@usgs.gov>
 License: License

{discontinuum-1.0.0 → discontinuum-1.0.1}/src/discontinuum/_version.py

@@ -17,5 +17,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '1.0.0'
-__version_tuple__ = version_tuple = (1, 0, 0)
+__version__ = version = '1.0.1'
+__version_tuple__ = version_tuple = (1, 0, 1)

{discontinuum-1.0.0 → discontinuum-1.0.1}/src/discontinuum.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: discontinuum
-Version: 1.0.0
+Version: 1.0.1
 Summary: Estimate discontinuous timeseries from continuous covariates.
 Maintainer-email: Timothy Hodson <thodson@usgs.gov>
 License: License

{discontinuum-1.0.0 → discontinuum-1.0.1}/src/rating_gp/providers/usgs.py

@@ -195,20 +195,50 @@ def get_measurements(
             NWISDischarge.column_name: NWISDischarge.standard_name,
             }
         )
-    # parse uncertainty from measured "measured_rating_diff"
-    qualitycode_to_uncertainty_fraction = {
+    
+    # Process the control_type_cd column
+    df["control_type_cd"] = (
+        df["control_type_cd"]
+        .fillna("Unspecified")
+        .astype("category")
+        )
+    
+    # Filter any measurements that are not used in the rating
+    mask = df["q_meas_used_fg"].str.lower().isin(['yes', 'y'])
+    df = df[mask]
+
+    num_not_used = (~mask).sum()
+    if num_not_used > 0:
+        warnings.warn(
+            f"{num_not_used} measurements were not used in the rating and "
+            "will be dropped from the dataset.",
+            UserWarning,
+        )
+
+    
+
+    # Quantitative values of "measured_rating_diff"
+    quality_codes = {
         'Excellent': '0.02',
         'Good': '0.05',
         'Fair': '0.08',
         'Poor': '0.12',
         'Unspecified': '0.12',
     }
+    # Replace other values with 'Unspecified'
+    df['measured_rating_diff'] = df['measured_rating_diff'].where(
+        df['measured_rating_diff'].isin(quality_codes),
+        'Unspecified'
+    )
+
+
     df['discharge_unc_frac'] = (df['measured_rating_diff']
-                                .replace(qualitycode_to_uncertainty_fraction)
+                                .replace(quality_codes)
                                 .astype(float))
-    # set indirect measurements as 20% uncertain regardless of quality code
+    # Set indirect measurements as 20% uncertain regardless of quality code
     df.loc[df['streamflow_method'] == 'QIDIR', 'discharge_unc_frac'] = 0.2
-    # convert fractional uncertainty to uncertainty assuming the uncertainty
+
+    # Convert fractional uncertainty to uncertainty assuming the uncertainty
     # fraction is a 2 sigma gse interval. (GSE = frac + 1)
     # (GSE -> exp(sigma_ln(Q)))
     df['discharge_unc'] = df['discharge_unc_frac'] / 2 + 1
@@ -216,7 +246,14 @@ def get_measurements(
     # drop data that is <= 0 as we need all positive data
     df = df[(df['stage'] > 0) & (df['discharge'] > 0)]
 
-    ds = xr.Dataset.from_dataframe(df[["stage", "discharge", "discharge_unc"]])
+    ds = xr.Dataset.from_dataframe(
+        df[[
+            "stage",
+            "discharge",
+            "discharge_unc",
+            "control_type_cd",
+        ]]
+        )
 
     for param in [NWISStage, NWISDischarge]:
         ds[param.name] = ds[param.name] * param.conversion