avoca 0.10.4__py3-none-any.whl → 0.11.1__py3-none-any.whl

avoca/bindings/ebas.py

@@ -17,50 +17,11 @@ from ebas.io.file.nasa_ames import EbasNasaAmes
 from nilutility.datatypes import DataObject
 from nilutility.datetime_helper import DatetimeInterval
 
+from avoca.bindings.ebas_flags import ebas_flag_to_avoca, flags_to_ebas
 from avoca.flags import QA_Flag
 
 logger = logging.getLogger(__name__)
 
-# https://projects.nilu.no/ccc/flags/flags.html for more info on what ebas uses
-flags_to_ebas: dict[QA_Flag, int] = {
-    QA_Flag.MISSING: 999,  # 	M 	Missing measurement, unspecified reason
-    QA_Flag.ZERO_NEG_CONC_EXT: 999,
-    QA_Flag.INVALIDATED_EXT: 900,  # 	H 	Hidden and invalidated by data originator
-    # V Extremely high value, outside four times standard deviation in a lognormal distribution
-    QA_Flag.EXTREME_VALUE: 458,
-    QA_Flag.CALIBRATION: 683,  # 	I 	Invalid due to calibration. Used for Level 0.
-    QA_Flag.BLANK: 684,  #  	Invalid due to zero/span check. Used for Level 0.
-    QA_Flag.HEIGHT_INTEGRATION: 0,  # 	Valid
-    QA_Flag.UNCORRELATED: 0,  # 	Valid
-    QA_Flag.MET_OFFICE_BASELINE: 0,  # 	Valid
-    QA_Flag.BELOW_DETECTION_LIMIT: 147,  # 	B 	Below detection limit
-    QA_Flag.POLLUTION: 900,
-    QA_Flag.SUSPICIOUS_RT: 900,
-    QA_Flag.INVALID_VALUES: 999,  # 	M 	Missing measurement, unspecified reason
-}
-
-ebas_flag_to_avoca: dict[int, QA_Flag] = {
-    ebas_flag: avoca_flag for avoca_flag, ebas_flag in flags_to_ebas.items()
-}
-# Set some flags with Multiple values to the same value
-ebas_flag_to_avoca.pop(0)  # 0 is valid in avoca
-ebas_flag_to_avoca[999] = QA_Flag.MISSING
-ebas_flag_to_avoca[900] = QA_Flag.INVALIDATED_EXT
-# Unspecified contamination or local influence, but considered valid
-ebas_flag_to_avoca[559] = QA_Flag.POLLUTION
-ebas_flag_to_avoca[685] = (
-    QA_Flag.CALIBRATION
-)  #  	Invalid due to secondary standard gas measurement. Used for Level 0.
-ebas_flag_to_avoca[980] = (
-    QA_Flag.CALIBRATION
-)  # Missing due to calibration or zero/span check
-
-missing_flags = set(QA_Flag) - set(flags_to_ebas.keys())
-if missing_flags:
-    raise RuntimeError(
-        f"Not all QA flags are mapped to Ebas flags. Missing: {missing_flags}"
-    )
-
 
 ebas_compname_of_var = {
     "rt": "retention_time",
@@ -192,7 +153,7 @@ def set_dataframe(
             serie_to_export = df_export[(sub, var)]
 
             values = [
-                None if pd.isna(val) or isnan else float(val)
+                None if pd.isna(val) or isnan or (not np.isfinite(val)) else float(val)
                 for val, isnan in zip(serie_to_export, nan_flag)
             ]
 

avoca/bindings/ebas_flags.py

@@ -0,0 +1,59 @@
+# https://projects.nilu.no/ccc/flags/flags.html for more info on what ebas uses
+from avoca.flags import QA_Flag
+
+flags_to_ebas: dict[QA_Flag, int] = {
+    QA_Flag.MISSING: 999,  # 	M 	Missing measurement, unspecified reason
+    QA_Flag.ZERO_NEG_CONC_EXT: 999,
+    QA_Flag.INVALIDATED_EXT: 900,  # 	H 	Hidden and invalidated by data originator
+    # V Extremely high value, outside four times standard deviation in a lognormal distribution
+    QA_Flag.EXTREME_VALUE: 458,
+    QA_Flag.CALIBRATION: 683,  # 	I 	Invalid due to calibration. Used for Level 0.
+    QA_Flag.BLANK: 684,  #  	Invalid due to zero/span check. Used for Level 0.
+    QA_Flag.HEIGHT_INTEGRATION: 0,  # 	Valid
+    QA_Flag.UNCORRELATED: 0,  # 	Valid
+    QA_Flag.MET_OFFICE_BASELINE: 0,  # 	Valid
+    QA_Flag.BELOW_DETECTION_LIMIT: 147,  # 	B 	Below detection limit
+    QA_Flag.POLLUTION: 900,
+    QA_Flag.SUSPICIOUS_RT: 900,
+    QA_Flag.INVALID_VALUES: 999,  # 	M 	Missing measurement, unspecified reason
+}
+
+ebas_flag_to_avoca: dict[int, QA_Flag] = {
+    ebas_flag: avoca_flag for avoca_flag, ebas_flag in flags_to_ebas.items()
+}
+# Set some flags with Multiple values to the same value
+ebas_flag_to_avoca.pop(0)  # 0 is valid in avoca
+ebas_flag_to_avoca[999] = QA_Flag.MISSING
+ebas_flag_to_avoca[900] = QA_Flag.INVALIDATED_EXT
+# Unspecified contamination or local influence, but considered valid
+ebas_flag_to_avoca[559] = QA_Flag.POLLUTION
+ebas_flag_to_avoca[685] = (
+    QA_Flag.CALIBRATION
+)  #  	Invalid due to secondary standard gas measurement. Used for Level 0.
+ebas_flag_to_avoca[980] = (
+    QA_Flag.CALIBRATION
+)  # Missing due to calibration or zero/span check
+
+missing_flags = set(QA_Flag) - set(flags_to_ebas.keys())
+if missing_flags:
+    raise RuntimeError(
+        f"Not all QA flags are mapped to Ebas flags. Missing: {missing_flags}"
+    )
+
+# priority of the flag to appear in the output
+# Useful when you can select only one flag value
+flag_order = [
+    QA_Flag.CALIBRATION,
+    QA_Flag.BLANK,
+    QA_Flag.HEIGHT_INTEGRATION,
+    QA_Flag.MET_OFFICE_BASELINE,
+    QA_Flag.BELOW_DETECTION_LIMIT,
+    QA_Flag.POLLUTION,
+    QA_Flag.SUSPICIOUS_RT,
+    QA_Flag.UNCORRELATED,
+    QA_Flag.EXTREME_VALUE,
+    QA_Flag.INVALIDATED_EXT,
+    QA_Flag.ZERO_NEG_CONC_EXT,
+    QA_Flag.MISSING,
+    QA_Flag.INVALID_VALUES,
+]

avoca/bindings/qa_tool.py

@@ -0,0 +1,209 @@
+"""Few modules for importing and exporting from https://voc-qc.nilu.no/
+
+Originally taken from tucavoc.
+"""
+
+import logging
+import warnings
+from datetime import datetime, timedelta
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+import pandas.errors
+
+from avoca.bindings.ebas_flags import flag_order, flags_to_ebas
+from avoca.flags import QA_Flag
+from avoca.utils import compounds_from_df
+
+
+def number_of_digits_required(serie: pd.Series) -> int:
+    """Return the number of digits required for the calculation"""
+    # TODO: need to check if we need the actual int  value, we can put a .9 at the end
+    if all(pd.isna(serie) | (serie == 0)):
+        # Only 2 will be required
+        return 2
+    else:
+        number_of_digits = np.log10(serie[serie > 0])
+        max_digits = number_of_digits[number_of_digits != np.inf]
+        if len(max_digits) == 0:
+            return 2
+        return int(max(np.max(max_digits), 0) + 2)
+
+
+def export_EmpaQATool(
+    df: pd.DataFrame,
+    export_path: Path,
+    station: str = "XXX",
+    revision_date: datetime | None = None,
+    dataset: datetime | str | None = None,
+    export_names: dict[str, str] = {},
+    datetime_offsets: tuple[timedelta, timedelta] | None = None,
+    substances: list[str] = [],
+    rounding_decimals: int = 4,
+) -> Path:
+    """Export to the EmpaQATool format.
+
+    The exported file from the program can then be imported to
+    the tool on https://voc-qc.nilu.no/Import
+    The specs fro that file can be found in
+    https://voc-qc.nilu.no/doc/CSVImport_FormatSpecifications.pdf
+
+    This will add the additional data from the dataframe.
+
+    The file genereated will be named:
+    export_path/[station]_[dataset]_[revision].csv
+
+    :arg df: Calculation dataframe
+    :arg export_path: Path (directory) to export the file
+    :arg station: Station name to use in the file name
+    :arg revision_date: Revision date as datetime to use in the file name
+    :arg dataset: Dataset name as datetime or string to use in the file name
+    :arg export_names: Dictionary of substance names to use in the file name
+        The keys are the substance names and the values are the names to use in the file.
+    :arg datetime_offsets: Tuple of two timedelta to use for the start and end datetime
+    :arg substances: List of substances to export. You can also specify group names.
+        If not specified, this will use the substances from `df_substances`.
+    :arg rounding_decimals: Number of decimals to round the values to.
+
+    """
+
+    logger = logging.getLogger(__name__)
+
+    warnings.filterwarnings(
+        action="ignore",
+        category=pandas.errors.PerformanceWarning,
+        module="pandas",
+    )
+
+    # fmt = "%Y-%m-%d %H:%M:%S"
+    fmt = "%d.%m.%Y %H:%M:%S"
+
+    need_datetime_col = ("-", "datetime_start") not in df.columns and (
+        "-",
+        "datetime_end",
+    ) not in df.columns
+
+    if need_datetime_col:
+        if ("-", "datetime") not in df.columns:
+            df[("-", "datetime")] = df.index
+        # Check type of the datetime column
+        if not pd.api.types.is_datetime64_any_dtype(df[("-", "datetime")]):
+            raise ValueError(
+                "The datetime column is not of type datetime64. "
+                "Please convert it to datetime64."
+                "Or provide ()"
+            )
+        if datetime_offsets is None:
+            raise ValueError(
+                "No datetime_start or datetime_end column in the dataframe. "
+                "Please provide the datetime_offsets to specify."
+            )
+
+        df[("-", "datetime_start")] = df[("-", "datetime")] + datetime_offsets[0]
+        df[("-", "datetime_end")] = df[("-", "datetime")] + datetime_offsets[1]
+
+    df_out = pd.DataFrame(
+        {
+            "start": df[("-", "datetime_start")].dt.strftime(fmt),
+            "end": df[("-", "datetime_end")].dt.strftime(fmt),
+        },
+        index=df.index,
+    )
+    logger.debug(f"df_out: {df_out.head()}")
+    if not substances:
+        substances = compounds_from_df(df)
+
+    remove_infs = lambda x: x.replace([np.inf, -np.inf], np.nan)
+    is_invalid = lambda x: x.isin([np.inf, -np.inf]) | pd.isna(x)
+    clean_col = lambda x: remove_infs(x).round(rounding_decimals).astype(str)
+
+    for substance in substances:
+
+        export_name = export_names.get(substance, substance)
+
+        conc_col = (
+            (substance, "conc")
+            if (substance, "conc") in df.columns
+            else (substance, "C")
+        )
+        u_expanded_col = (substance, "u_expanded")
+        u_precision_col = (substance, "u_precision")
+        flag_col = (substance, "flag")
+
+        mask_invalid = (
+            (
+                df[flag_col] & (QA_Flag.MISSING.value + QA_Flag.INVALIDATED_EXT.value)
+            ).astype(bool)
+            | is_invalid(df[conc_col])
+            | (
+                is_invalid(df[u_expanded_col])
+                if u_expanded_col in df.columns
+                else False
+            )
+            | (
+                is_invalid(df[u_precision_col])
+                if u_precision_col in df.columns
+                else False
+            )
+        )
+
+        logger.debug(f"mask_invalid: {mask_invalid}")
+        # Flag the invalids
+        df.loc[mask_invalid, flag_col] ^= QA_Flag.INVALID_VALUES.value
+
+        # Convert to str so we can control the formatting
+        df_out[f"{export_name}-Value"] = clean_col(df[conc_col])
+
+        # Input the missing values as 9. see issue #7 gitlab.empa.ch
+        df_out.loc[mask_invalid, f"{export_name}-Value"] = (
+            "9" * number_of_digits_required(df[conc_col])
+        )
+
+        if u_expanded_col in df.columns:
+            # Convert to str so we can control the formatting
+            df_out[f"{export_name}-Accuracy"] = clean_col(df[u_expanded_col])
+            # Input the missing values as 9. see issue #7 gitlab.empa.ch
+            df_out.loc[mask_invalid, f"{export_name}-Accuracy"] = (
+                "9" * number_of_digits_required(df[u_expanded_col])
+            )
+
+        if u_precision_col in df.columns:
+            # Convert to str so we can control the formatting
+            df_out[f"{export_name}-Precision"] = clean_col(df[u_precision_col])
+
+            # Input the missing values as 9. see issue #7 gitlab.empa.ch
+            df_out.loc[mask_invalid, f"{export_name}-Precision"] = (
+                "9" * number_of_digits_required(df[u_precision_col])
+            )
+
+        flag_col_out = f"{export_name}-Flag"
+        df_out[flag_col_out] = 0.0
+        for flag in flag_order:
+            df_out.loc[
+                (df[flag_col].values & flag.value).astype(bool), flag_col_out
+            ] = flags_to_ebas[flag]
+        df_out[flag_col_out] = (df_out[flag_col_out] * 1e-3).map("{:.3f}".format)
+
+    export_path.mkdir(exist_ok=True)
+
+    dt_format = "%Y%m%d"
+    if dataset is None:
+        dataset = datetime.strptime(df_out["start"].iloc[0], fmt).strftime(dt_format)
+
+    if revision_date is None:
+        revision_date = datetime.now().strftime(dt_format)
+
+    # [station]_[dataset]_[revision]
+    file_name = f"{station}_{dataset}_{revision_date}"
+
+    out_filepath = Path(export_path, file_name).with_suffix(".csv")
+    df_out.to_csv(
+        out_filepath,
+        sep=";",
+        index=False,
+        encoding="utf-8",
+    )
+    logger.info(f"Exported to `{out_filepath}`")
+
+    return out_filepath

avoca/testing/df.py

@@ -23,6 +23,16 @@ simple_df = pd.DataFrame(
     ),
 )
 
+invalids_df = pd.DataFrame(
+    np.transpose([[1.0, 1.1, 0.8, 0.9], [1.0, np.inf, np.nan, -0.3]]),
+    columns=pd.MultiIndex.from_tuples(
+        [
+            ("compA", "C"),
+            ("compB", "C"),
+        ]
+    ),
+)
+
 compab_multiindex = pd.MultiIndex.from_tuples(
     [
         ("compA", "test_var"),

{avoca-0.10.4.dist-info → avoca-0.11.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: avoca
-Version: 0.10.4
+Version: 0.11.1
 Summary: @voc@: Quality assessement of measurement data
 Project-URL: Homepage, https://gitlab.com/empa503/atmospheric-measurements/avoca
 Project-URL: Bug Tracker, https://gitlab.com/empa503/atmospheric-measurements/avoca/-/issues

{avoca-0.10.4.dist-info → avoca-0.11.1.dist-info}/RECORD

@@ -8,11 +8,13 @@ avoca/manager.py,sha256=ET-ATrSLi2rSV7PjBzwpjj0V_60MFxSIZqQ03aEIbdA,5284
 avoca/requirements.py,sha256=q4z6bJ6iW5jSy10Y0elfE9BoEcAZC2-kUqYi4zA6TGE,563
 avoca/settings.py,sha256=Px-sCGIlRyWI2RBJaGUY0K1V60kOZY9n41eft92bjN4,2112
 avoca/bindings/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-avoca/bindings/ebas.py,sha256=U5RhjQuEvYcEQQ17XDj2HOQ_Wr1WX03aR_kL3BVmsO8,18975
+avoca/bindings/ebas.py,sha256=DeIRhHbxj76TMy3Z_Ok2AgiGzSkb6pW7scIMORgh3y0,17243
+avoca/bindings/ebas_flags.py,sha256=ls8cEKPC2QjlkWyqiytyjFfemqM8ot8suys_Qi5Xx1o,2352
 avoca/bindings/gcwerks-report.conf,sha256=jO0I62DfgzrXXS1FuiW8ds-oc1_j8kpFCO61Fk-erBw,230
 avoca/bindings/gcwerks.py,sha256=pTwZhSuoD4usER1-JhQJtgj1KUcZR1ZN1loZMCSd3TQ,14651
 avoca/bindings/gcwerks_gui.py,sha256=Fj3p8obFq3lWrWW0LlA8WBALP8-U70hvps5vZEt4NaM,9458
 avoca/bindings/nabel.py,sha256=VbC_ARvtso8onILAD8gROt5Y2URdx6NfAqMn4p1mUWU,3020
+avoca/bindings/qa_tool.py,sha256=ZPtQo8dHo6wooIlc9Vzk8y91Qgso-RBtGR_h2TAZQ24,7583
 avoca/bindings/synspec.py,sha256=W5RnBu-6eetmwjM8iMBe4wNwVNIaVpNW3bwa2ykGM2U,1733
 avoca/qa_class/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 avoca/qa_class/abstract.py,sha256=4s8GgkeC3WbMnNxygajhawh7TU9v13i-SP0j5b5YOMc,5432
@@ -23,11 +25,11 @@ avoca/qa_class/rt.py,sha256=t927H_o0Kn-VwEkG9TW33MSDW_2of2-pBX_gTMuaXA0,3817
 avoca/qa_class/test.py,sha256=Xc88_Vwf3hvPiKKl4ILxZ2N985SY8eujUdnAoQu4mbo,591
 avoca/qa_class/zscore.py,sha256=HqOxV45smhXqcv2XrB7W7plE9RoHzBGVEAbmuwsiv7w,16696
 avoca/testing/__init__.py,sha256=CzkugadVit48-eMoMVtojZLHeSKgnmMMen6sGu6Q42Y,108
-avoca/testing/df.py,sha256=D1ONXe2b6vuM68sTIh318dcRRqodQ5KQioiVMl6OJYo,1592
+avoca/testing/df.py,sha256=Nc0GUYTApZgYyUTMnHMTbSKLiA5ty9Bg7gUGtnoFYMI,1826
 avoca/utils/__init__.py,sha256=LEA2jJsqwSK2DBzXg00DbPhM1fXXREJ0XxLeuJtKapY,1398
 avoca/utils/flags_doc.py,sha256=cS7yKpxVh_SA6EdH3lSy9UpcIvhGwzAELRbkXN3CxO8,4168
 avoca/utils/torch_models.py,sha256=53TgOgSPMOOSGYy2cm1EGSK7qQkYMGEOq319KKM_Ir0,1015
-avoca-0.10.4.dist-info/METADATA,sha256=usLBL-Kyms4faFat9-TYcYtw-U-jSqFDgMCQeZCFdjc,1570
-avoca-0.10.4.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-avoca-0.10.4.dist-info/licenses/LICENCE.txt,sha256=4MY53j3v7tEKwjyuriVz9YjB4Dscm2nDMB2CcG9lOmk,1059
-avoca-0.10.4.dist-info/RECORD,,
+avoca-0.11.1.dist-info/METADATA,sha256=F6Au_DcqiIbQFSIGBDpqUt76yHMsZF0MX6doqEYPpxA,1570
+avoca-0.11.1.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+avoca-0.11.1.dist-info/licenses/LICENCE.txt,sha256=4MY53j3v7tEKwjyuriVz9YjB4Dscm2nDMB2CcG9lOmk,1059
+avoca-0.11.1.dist-info/RECORD,,