discontinuum 1.0.0__py3-none-any.whl → 1.0.2__py3-none-any.whl

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 = '1.0.2'
+__version_tuple__ = version_tuple = (1, 0, 2)

{discontinuum-1.0.0.dist-info → discontinuum-1.0.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: discontinuum
-Version: 1.0.0
+Version: 1.0.2
 Summary: Estimate discontinuous timeseries from continuous covariates.
 Maintainer-email: Timothy Hodson <thodson@usgs.gov>
 License: License

{discontinuum-1.0.0.dist-info → discontinuum-1.0.2.dist-info}/RECORD

@@ -1,5 +1,5 @@
 discontinuum/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-discontinuum/_version.py,sha256=fo5PXsZuloQZu3LdpIFTUAXvJmY2L9N5sNGe2tvdU98,511
+discontinuum/_version.py,sha256=98Mtoja-WpqDUnppClywkGhX6oxX6l-ZHnSYMYYbBUE,511
 discontinuum/data_manager.py,sha256=LiZoPR0nnu7YAUfh5L1ZDRfaS3dgfVIELXIHkzUKyBg,4416
 discontinuum/pipeline.py,sha256=1avuZnFai-b3HmihcpZ8M3WFNQ8lXAFSNTrnfl2NrY0,10074
 discontinuum/plot.py,sha256=eZQS6-Ydq8FFcEukPtNuDVB-weV6lHyWMyJ1hqTkVrU,2969
@@ -11,7 +11,7 @@ discontinuum/engines/pymc.py,sha256=phbtE-3UCSVcP1MhbXwAHIWDZWDr56wK9U7aRt-w-2o,
 discontinuum/providers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 discontinuum/providers/base.py,sha256=Yn2EHS1b4fYl09-m2MYuf2P9VRUXAP-WDpSoZrCbRvY,720
 discontinuum/tests/test_pipeline.py,sha256=_FhkGxbFIxNb35lGaIdZk7Zjgs6CkxEF3gFUX3PE8EU,918
-discontinuum-1.0.0.dist-info/licenses/LICENSE.md,sha256=XElVHHnS2uQ15M_Z2giPH1vmeWMzdpGQ48ItkuZurVA,1650
+discontinuum-1.0.2.dist-info/licenses/LICENSE.md,sha256=XElVHHnS2uQ15M_Z2giPH1vmeWMzdpGQ48ItkuZurVA,1650
 loadest_gp/__init__.py,sha256=YISfvbc7Zy2y0BOxS1A2KzqxyoNJTz0EnLMnRW6iVT8,740
 loadest_gp/plot.py,sha256=x2PK7vBCc44dX9lu5YV-rvw1u4pvXSLdcrTSvYLiHMA,2595
 loadest_gp/utils.py,sha256=m5QaqR_0JiuRXPfryH8nI5lODp8PqvQla5C05WDN3LY,2772
@@ -28,8 +28,8 @@ rating_gp/models/base.py,sha256=e2Kq644I88YLHWPNA0qyRgitF5wimdLW4618vKX-o_s,1474
 rating_gp/models/gpytorch.py,sha256=rSxuTMoSeGK2LwqANMDMrmxDsOCXyAWSmENm6KFjRZ0,5930
 rating_gp/models/kernels.py,sha256=3xg2mhY3aEgjI3r5vyAll9MA4c3M5UKqRi3FApNhJJQ,11579
 rating_gp/providers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-rating_gp/providers/usgs.py,sha256=I3iilF5jHk3DOejD_qoSyR-nY70GGjRI4Aw8yIMNhmQ,6094
-discontinuum-1.0.0.dist-info/METADATA,sha256=sT0RWAmZ683JHKq3363yE43-VpJ0G12_vUlxjg4mcQg,6231
-discontinuum-1.0.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-discontinuum-1.0.0.dist-info/top_level.txt,sha256=mwU_PSFrZYSJrBgqIuTJTo7Pp9ODDv6XdDed7kAagXM,34
-discontinuum-1.0.0.dist-info/RECORD,,
+rating_gp/providers/usgs.py,sha256=KmKYN3c8Mi-ly2l6X80WT3taEhqCPXeEcRNi9HvbJmY,8134
+discontinuum-1.0.2.dist-info/METADATA,sha256=0oCct8VOrQib5sCzdi2cBNI8KEcpYTYTKj1oCL3DCsc,6231
+discontinuum-1.0.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+discontinuum-1.0.2.dist-info/top_level.txt,sha256=mwU_PSFrZYSJrBgqIuTJTo7Pp9ODDv6XdDed7kAagXM,34
+discontinuum-1.0.2.dist-info/RECORD,,

rating_gp/providers/usgs.py

@@ -21,6 +21,16 @@ if TYPE_CHECKING:
 FT_TO_M = 0.3048
 FT3_TO_M3 = 0.0283168
 
+    # Quantitative values of "measured_rating_diff"
+USGS_QUALITY_CODES = {
+    'Excellent': '0.02',
+    'Good': '0.05',
+    'Fair': '0.08',
+    'Poor': '0.12',
+    'Unspecified': '0.12',
+}
+
+
 @dataclass
 class NWISColumn:
     column_name: str
@@ -182,6 +192,53 @@ def get_measurements(
     )
     df = nwis._read_rdb(response.text)
 
+    return read_measurements_df(df)    
+
+
+def read_measurements_df(df: pd.DataFrame) -> xr.Dataset:
+    """Read a DataFrame of USGS discharge measurements and convert to xarray Dataset.
+
+    Parameters
+    ----------
+    df : pd.DataFrame
+        Dataframe from `dataretrieval.nwis.get_discharge_measurements()`
+
+    Returns
+    -------
+    xr.Dataset
+    
+    Example
+    -------
+    >>> from dataretrieval import nwis
+    >>> from rating_gp.providers.usgs import read_measurements_df
+    >>> df, _ = nwis.get_discharge_measurements(
+        sites='03339000',
+        start='2020-01-01',
+        end='2020-12-31',
+        format='rdb_expanded',
+        )
+    >>> ds = read_measurements_df(df)
+    """
+
+    # assert the correct columns are present
+    required_columns = [
+        "measurement_dt",
+        "gage_height_va",
+        "discharge_va",
+        "q_meas_used_fg",
+        "control_type_cd",
+        "measured_rating_diff",
+        "streamflow_method",
+        NWISStage.column_name,
+        NWISDischarge.column_name,
+    ]
+
+    missing_columns = set(required_columns) - set(df.columns)
+    if missing_columns:
+        raise ValueError(
+            f"Missing required columns in the DataFrame: {missing_columns}"
+        )
+
     # covert timezone to UTC? ignore for now
     df.index = pd.to_datetime(
         df["measurement_dt"],
@@ -195,20 +252,40 @@ def get_measurements(
             NWISDischarge.column_name: NWISDischarge.standard_name,
             }
         )
-    # parse uncertainty from measured "measured_rating_diff"
-    qualitycode_to_uncertainty_fraction = {
-        'Excellent': '0.02',
-        'Good': '0.05',
-        'Fair': '0.08',
-        'Poor': '0.12',
-        'Unspecified': '0.12',
-    }
+    
+    # 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,
+        )
+
+    # Replace other values with 'Unspecified'
+    df['measured_rating_diff'] = df['measured_rating_diff'].where(
+        df['measured_rating_diff'].isin(USGS_QUALITY_CODES.keys()),
+        'Unspecified'
+    )
+
     df['discharge_unc_frac'] = (df['measured_rating_diff']
-                                .replace(qualitycode_to_uncertainty_fraction)
+                                .replace(USGS_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 +293,14 @@ 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