pydartdiags 0.0.42__py3-none-any.whl → 0.0.43__py3-none-any.whl

pydartdiags/obs_sequence/obs_sequence.py

@@ -5,6 +5,23 @@ import os
 import yaml
 import struct
 
+def requires_assimilation_info(func):
+    def wrapper(self, *args, **kwargs):
+        if self.has_assimilation_info:
+            return func(self, *args, **kwargs)
+        else:
+            raise ValueError("Assimilation information is required to call this function.")
+    return wrapper
+
+def requires_posterior_info(func):
+    def wrapper(self, *args, **kwargs):
+        if self.has_posterior_info:
+            return func(self, *args, **kwargs)
+        else:
+            raise ValueError("Posterior information is required to call this function.")
+    return wrapper
+
+
 class obs_sequence:
     """Create an obs_sequence object from an ascii observation sequence file.
 
@@ -59,6 +76,8 @@ class obs_sequence:
     
     def __init__(self, file, synonyms=None):
         self.loc_mod = 'None'
+        self.has_assimilation_info = False
+        self.has_posterior = False
         self.file = file
         self.synonyms_for_obs = ['NCEP BUFR observation',
                                  'AIRS observation', 
@@ -72,6 +91,17 @@ class obs_sequence:
             else:
                 self.synonyms_for_obs.append(synonyms)
 
+        if file is None:
+            # Early exit for testing purposes
+            self.df = pd.DataFrame()
+            self.types = {}
+            self.reverse_types = {}
+            self.copie_names = []
+            self.n_copies = 0
+            self.seq = []
+            self.all_obs = []
+            return
+
         module_dir = os.path.dirname(__file__)
         self.default_composite_types = os.path.join(module_dir,"composite_types.yaml")
 
@@ -103,11 +133,16 @@ class obs_sequence:
         self.synonyms_for_obs = [synonym.replace(' ', '_') for synonym in self.synonyms_for_obs]
         rename_dict = {old: 'observation' for old in self.synonyms_for_obs  if old in self.df.columns}
         self.df = self.df.rename(columns=rename_dict)
+
         # calculate bias and sq_err is the obs_seq is an obs_seq.final
         if 'prior_ensemble_mean'.casefold() in map(str.casefold, self.columns):
-            self.df['bias'] = (self.df['prior_ensemble_mean'] - self.df['observation'])
-            self.df['sq_err'] = self.df['bias']**2  # squared error
-	
+            self.has_assimilation_info = True
+            self.df['prior_bias'] = (self.df['prior_ensemble_mean'] - self.df['observation'])
+            self.df['prior_sq_err'] = self.df['prior_bias']**2  # squared error
+        if 'posterior_ensemble_mean'.casefold() in map(str.casefold, self.columns):
+            self.has_posterior_info = True
+            self.df['posterior_bias'] = (self.df['posterior_ensemble_mean'] - self.df['observation'])
+            self.df['posterior_sq_err'] = self.df['posterior_bias']**2
 
     def create_all_obs(self):
         """ steps through the generator to create a
@@ -152,14 +187,38 @@ class obs_sequence:
             data.append(self.types[type_value]) # observation type
             
         # any observation specific obs def info is between here and the end of the list
+        # can be obs_def & external forward operator
+        metadata = obs[typeI+2:-2]
+        obs_def_metadata, external_metadata = self.split_metadata(metadata)
+        data.append(obs_def_metadata)
+        data.append(external_metadata)
+
         time = obs[-2].split()
         data.append(int(time[0])) # seconds
         data.append(int(time[1])) # days
         data.append(convert_dart_time(int(time[0]), int(time[1]))) # datetime   # HK todo what is approprate for 1d models?
         data.append(float(obs[-1])) # obs error variance ?convert to sd?
-        
+
         return data
 
+    @staticmethod
+    def split_metadata(metadata):
+        """
+        Split the metadata list at the first occurrence of an element starting with 'externalF0'.
+
+        Args:
+            metadata (list of str): The metadata list to be split.
+
+        Returns:
+            tuple: Two sublists, the first containing elements before 'externalF0', and the second
+                containing 'externalF0' and all elements after it. If 'externalF0' is not found,
+                the first sublist contains the entire metadata list, and the second is empty.
+        """
+        for i, item in enumerate(metadata):
+            if item.startswith('external_FO'):
+                return metadata[:i], metadata[i:]
+        return metadata, []
+
     def list_to_obs(self, data):
         obs = []
         obs.append('OBS        ' + str(data[0]))  # obs_num lots of space
@@ -171,10 +230,16 @@ class obs_sequence:
             obs.append('   '.join(map(str, data[self.n_copies+2:self.n_copies+5])) + '   ' + str(self.reversed_vert[data[self.n_copies+5]]) )  # location x, y, z, vert
             obs.append('kind') # this is type of observation
             obs.append(self.reverse_types[data[self.n_copies + 6]])  # observation type
+            # Convert metadata to a string and append
+            obs.extend(data[self.n_copies + 7])  # metadata
         elif self.loc_mod == 'loc1d':
             obs.append(data[self.n_copies+2])  # 1d location
             obs.append('kind') # this is type of observation
             obs.append(self.reverse_types[data[self.n_copies + 3]])  # observation type
+            # Convert metadata to a string and append
+            metadata = ' '.join(map(str, data[self.n_copies + 4:-4]))
+            if metadata:
+                obs.append(metadata)  # metadata
         obs.append(' '.join(map(str, data[-4:-2])))  # seconds, days
         obs.append(data[-1])  # obs error variance
 
@@ -273,12 +338,70 @@ class obs_sequence:
         elif self.loc_mod == 'loc1d':
             heading.append('location')
         heading.append('type')
+        heading.append('metadata')
+        heading.append('external_FO')
         heading.append('seconds')
         heading.append('days')
         heading.append('time')
         heading.append('obs_err_var')
         return heading
 
+    @requires_assimilation_info    
+    def select_by_dart_qc(self, dart_qc):
+        """
+        Selects rows from a DataFrame based on the DART quality control flag.
+
+        Parameters:
+            df (DataFrame): A pandas DataFrame.
+            dart_qc (int): The DART quality control flag to select.
+
+        Returns:
+            DataFrame: A DataFrame containing only the rows with the specified DART quality control flag.
+
+        Raises:
+            ValueError: If the DART quality control flag is not present in the DataFrame.
+        """
+        if dart_qc not in self.df['DART_quality_control'].unique():
+            raise ValueError(f"DART quality control flag '{dart_qc}' not found in DataFrame.")
+        else:
+            return self.df[self.df['DART_quality_control'] == dart_qc]
+
+    @requires_assimilation_info
+    def select_failed_qcs(self):
+        """
+        Select rows from the DataFrame where the DART quality control flag is greater than 0.
+
+        Returns:
+            pandas.DataFrame: A DataFrame containing only the rows with a DART quality control flag greater than 0.
+        """
+        return self.df[self.df['DART_quality_control'] > 0]
+
+    @requires_assimilation_info
+    def possible_vs_used(self):
+        """
+        Calculates the count of possible vs. used observations by type.
+
+        This function takes a DataFrame containing observation data, including a 'type' column for the observation
+        type and an 'observation' column. The number of used observations ('used'), is the total number
+        minus the observations that failed quality control checks (as determined by the `select_failed_qcs` function).
+        The result is a DataFrame with each observation type, the count of possible observations, and the count of
+        used observations.
+
+        Returns:
+            pd.DataFrame: A DataFrame with three columns: 'type', 'possible', and 'used'. 'type' is the observation type,
+            'possible' is the count of all observations of that type, and 'used' is the count of observations of that type
+            that passed quality control checks.
+        """
+        possible = self.df.groupby('type')['observation'].count()
+        possible.rename('possible', inplace=True)
+        
+        failed_qcs = self.select_failed_qcs().groupby('type')['observation'].count()
+        used = possible - failed_qcs.reindex(possible.index, fill_value=0)
+        used.rename('used', inplace=True)
+        
+        return pd.concat([possible, used], axis=1).reset_index()
+
+
     @staticmethod
     def is_binary(file):
         """Check if a file is binary file."""
@@ -659,65 +782,6 @@ def convert_dart_time(seconds, days):
     """
     time = dt.datetime(1601,1,1) + dt.timedelta(days=days, seconds=seconds)
     return time
-    
-def select_by_dart_qc(df, dart_qc):
-    """
-    Selects rows from a DataFrame based on the DART quality control flag.
-
-    Parameters:
-        df (DataFrame): A pandas DataFrame.
-        dart_qc (int): The DART quality control flag to select.
-
-    Returns:
-        DataFrame: A DataFrame containing only the rows with the specified DART quality control flag.
-
-    Raises:
-        ValueError: If the DART quality control flag is not present in the DataFrame.
-    """
-    if dart_qc not in df['DART_quality_control'].unique():
-        raise ValueError(f"DART quality control flag '{dart_qc}' not found in DataFrame.")
-    else:
-        return df[df['DART_quality_control'] == dart_qc]
-
-def select_failed_qcs(df):
-    """
-    Selects rows from a DataFrame where the DART quality control flag is greater than 0.
-
-    Parameters:
-        df (DataFrame): A pandas DataFrame.
-
-    Returns:
-        DataFrame: A DataFrame containing only the rows with a DART quality control flag greater than 0.
-    """
-    return df[df['DART_quality_control'] > 0]
-
-def possible_vs_used(df):
-    """
-    Calculates the count of possible vs. used observations by type.
-
-    This function takes a DataFrame containing observation data, including a 'type' column for the observation
-    type and an 'observation' column. The number of used observations ('used'), is the total number
-    minus the observations that failed quality control checks (as determined by the `select_failed_qcs` function).
-    The result is a DataFrame with each observation type, the count of possible observations, and the count of
-    used observations.
-
-    Parameters:
-        df (pd.DataFrame): A DataFrame with at least two columns: 'type' for the observation type and 'observation'
-        for the observation data. It may also contain other columns required by the `select_failed_qcs` function
-        to determine failed quality control checks.
-
-    Returns:
-        pd.DataFrame: A DataFrame with three columns: 'type', 'possible', and 'used'. 'type' is the observation type,
-        'possible' is the count of all observations of that type, and 'used' is the count of observations of that type
-        that passed quality control checks.
-
-    """
-    possible = df.groupby('type')['observation'].count()
-    possible.rename('possible', inplace=True)
-    used = df.groupby('type')['observation'].count() - select_failed_qcs(df).groupby('type')['observation'].count()
-    used.rename('used', inplace=True)
-    return pd.concat([possible, used], axis=1).reset_index()
-
 
 def construct_composit(df_comp, composite, components):
     """

pydartdiags/plots/plots.py

@@ -1,6 +1,7 @@
 
 import numpy as np
 import plotly.express as px
+import plotly.graph_objects as go
 import pandas as pd
 
 def plot_rank_histogram(df):  
@@ -62,55 +63,208 @@ def calculate_rank(df):
 
     return (rank, ens_size, result_df)
 
-def plot_profile(df, levels):
+def plot_profile(df, levels, verticalUnit = "pressure (Pa)"):
     """
-    Plots RMSE and Bias profiles for different observation types across specified pressure levels.
+    Plots RMSE, bias, and total spread profiles for different observation types across specified vertical levels.
 
     This function takes a DataFrame containing observational data and model predictions, categorizes
-    the data into specified pressure levels, and calculates the RMSE and Bias for each level and
-    observation type. It then plots two line charts: one for RMSE and another for Bias, both as functions
-    of pressure level. The pressure levels are plotted on the y-axis in reversed order to represent
-    the vertical profile in the atmosphere correctly.
+    the data into specified vertical levels, and calculates the RMSE, bias and total spread for each level and
+    observation type. It then plots three line charts: one for RMSE, one for bias, one for total spread, as functions
+    of vertical level. The vertical levels are plotted on the y-axis in reversed order to represent
+    the vertical profile in the atmosphere correctly if the vertical units are pressure.
 
     Parameters:
-        df (pd.DataFrame): The input DataFrame containing at least the 'vertical' column for pressure levels,
-        and other columns required by the `rmse_bias` function for calculating RMSE and Bias.
-        levels (array-like): The bin edges for categorizing the 'vertical' column values into pressure levels.
+        df (pd.DataFrame): The input DataFrame containing at least the 'vertical' column for vertical levels,
+        the vert_unit column, and other columns required by the `rmse_bias` function for calculating RMSE and 
+        Bias.
+        levels (array-like): The bin edges for categorizing the 'vertical' column values into the desired 
+        vertical levels.
+        verticalUnit (string) (optional): The vertical unit to be used. Only observations in df which have this
+        string in the vert_unit column will be plotted. Defaults to 'pressure (Pa)'.
 
     Returns:
-        tuple: A tuple containing the DataFrame with RMSE and Bias calculations, the RMSE plot figure, and the
-        Bias plot figure. The DataFrame includes a 'plevels' column representing the categorized pressure levels
-        and 'hPa' column representing the midpoint of each pressure level bin.
+        tuple: A tuple containing the DataFrame with RMSE, bias and total spread calculations,
+        The DataFrame includes a 'vlevels' column representing the categorized vertical levels
+        and 'midpoint' column representing the midpoint of each vertical level bin. And the three figures.
 
     Raises:
         ValueError: If there are missing values in the 'vertical' column of the input DataFrame.
+        ValueError: If none of the input obs have 'verticalUnit' in the 'vert_unit' column of the input DataFrame.
 
     Note:
-        - The function modifies the input DataFrame by adding 'plevels' and 'hPa' columns.
-        - The 'hPa' values are calculated as half the midpoint of each pressure level bin, which may need
-          adjustment based on the specific requirements for pressure level representation.
+        - The function modifies the input DataFrame by adding 'vlevels' and 'midpoint' columns.
+        - The 'midpoint' values are calculated as half the midpoint of each vertical level bin, which may need
+          adjustment based on the specific requirements for vertical level representation.
         - The plots are generated using Plotly Express and are displayed inline. The y-axis of the plots is
-          reversed to align with standard atmospheric pressure level representation.
+          reversed to align with standard atmospheric pressure level representation if the vertical units
+          are atmospheric pressure.
     """
 
     pd.options.mode.copy_on_write = True
     if df['vertical'].isnull().values.any(): # what about horizontal observations?
         raise ValueError("Missing values in 'vertical' column.")
+    elif verticalUnit not in df['vert_unit'].values:
+        raise ValueError("No obs with expected vertical unit '"+verticalUnit+"'.")
     else:
-        df.loc[:,'plevels'] = pd.cut(df['vertical'], levels)
-        df.loc[:,'hPa'] = df['plevels'].apply(lambda x: x.mid / 1000.) # HK todo units
+        df = df[df["vert_unit"].isin({verticalUnit})] # Subset to only rows with the correct vertical unit
+        df.loc[:,'vlevels'] = pd.cut(df['vertical'], levels)
+        if verticalUnit == "pressure (Pa)":
+            df.loc[:,'midpoint'] = df['vlevels'].apply(lambda x: x.mid / 100.) # HK todo units
+        else:
+            df.loc[:,'midpoint'] = df['vlevels'].apply(lambda x: x.mid)
 
-    df_profile = rmse_bias(df)
-    fig_rmse = px.line(df_profile, y='hPa', x='rmse', title='RMSE by Level', markers=True, color='type', width=800, height=800)
-    fig_rmse.update_yaxes(autorange="reversed")
-    fig_rmse.show()
+    # Calculations
+    df_profile_prior = rmse_bias_totalspread(df, phase='prior')
+    df_profile_posterior = None
+    if 'posterior_ensemble_mean' in df.columns:
+        df_profile_posterior = rmse_bias_totalspread(df, phase='posterior')
 
-    fig_bias = px.line(df_profile, y='hPa', x='bias', title='Bias by Level', markers=True, color='type', width=800, height=800)
-    fig_bias.update_yaxes(autorange="reversed")
-    fig_bias.show()
+    # Merge prior and posterior dataframes
+    if df_profile_posterior is not None:
+        df_profile = pd.merge(df_profile_prior, df_profile_posterior, on=['midpoint', 'type'], suffixes=('_prior', '_posterior'))
+        fig_rmse = plot_profile_prior_post(df_profile, 'rmse', verticalUnit)
+        fig_rmse.show()
+        fig_bias = plot_profile_prior_post(df_profile, 'bias', verticalUnit)
+        fig_bias.show()
+        fig_ts = plot_profile_prior_post(df_profile, 'totalspread', verticalUnit)
+        fig_ts.show()
+    else:
+        df_profile = df_profile_prior
+        fig_rmse = plot_profile_prior(df_profile, 'rmse', verticalUnit)
+        fig_rmse.show()
+        fig_bias = plot_profile_prior(df_profile, 'bias', verticalUnit)
+        fig_bias.show()
+        fig_ts = plot_profile_prior(df_profile, 'totalspread', verticalUnit)
+        fig_ts.show()   
 
-    return df_profile, fig_rmse, fig_bias
+    return df_profile, fig_rmse, fig_ts, fig_bias
     
+def plot_profile_prior_post(df_profile, stat, verticalUnit):
+    """
+    Plots prior and posterior statistics by vertical level for different observation types.
+
+    Parameters:
+        df_profile (pd.DataFrame): DataFrame containing the prior and posterior statistics.
+        stat (str): The statistic to plot (e.g., 'rmse', 'bias', 'totalspread').
+        verticalUnit (str): The unit of the vertical axis (e.g., 'pressure (Pa)').
+
+    Returns:
+        plotly.graph_objects.Figure: The generated Plotly figure.
+    """
+    # Reshape DataFrame to long format for easier plotting
+    df_long = pd.melt(
+        df_profile,
+        id_vars=["midpoint", "type"],
+        value_vars=["prior_"+stat, "posterior_"+stat],
+        var_name=stat+"_type",
+        value_name=stat+"_value"
+    )
+
+    # Define a color mapping for observation each type
+    unique_types = df_long["type"].unique()
+    colors = px.colors.qualitative.Plotly
+    color_mapping = {type_: colors[i % len(colors)] for i, type_ in enumerate(unique_types)}
+
+    # Create a mapping for line styles based on stat
+    line_styles = {"prior_"+stat: "solid", "posterior_"+stat: "dash"}
+
+    # Create the figure
+    fig_stat = go.Figure()
+
+    # Loop through each type and type to add traces
+    for t in df_long["type"].unique():
+        for stat_type, dash_style in line_styles.items():
+            # Filter the DataFrame for this type and stat
+            df_filtered = df_long[(df_long[stat+"_type"] == stat_type) & (df_long["type"] == t)]
+            
+            # Add a trace
+            fig_stat.add_trace(go.Scatter(
+                x=df_filtered[stat+"_value"],
+                y=df_filtered["midpoint"],
+                mode='lines+markers',
+                name='prior '+t if stat_type == "prior_"+stat else 'post ',  # Show legend for "prior_stat OBS TYPE" only
+                line=dict(dash=dash_style, color=color_mapping[t]),  # Same color for all traces in group
+                marker=dict(size=8, color=color_mapping[t]),
+                legendgroup=t  # Group traces by type
+            ))
+
+        # Update layout
+        fig_stat.update_layout(
+            title= stat+' by Level',
+            xaxis_title=stat,
+            yaxis_title=verticalUnit,
+            width=800,
+            height=800,
+            template="plotly_white"
+        )
+
+    if verticalUnit == "pressure (Pa)":
+        fig_stat.update_yaxes(autorange="reversed")
+
+    return fig_stat
+
+
+def plot_profile_prior(df_profile, stat, verticalUnit):
+    """
+    Plots prior statistics by vertical level for different observation types.
+
+    Parameters:
+        df_profile (pd.DataFrame): DataFrame containing the prior statistics.
+        stat (str): The statistic to plot (e.g., 'rmse', 'bias', 'totalspread').
+        verticalUnit (str): The unit of the vertical axis (e.g., 'pressure (Pa)').
+
+    Returns:
+        plotly.graph_objects.Figure: The generated Plotly figure.
+    """
+    # Reshape DataFrame to long format for easier plotting - not needed for prior only, but
+    #   leaving it in for consistency with the plot_profile_prior_post function for now
+    df_long = pd.melt(
+        df_profile,
+        id_vars=["midpoint", "type"],
+        value_vars=["prior_"+stat],
+        var_name=stat+"_type",
+        value_name=stat+"_value"
+    )
+
+    # Define a color mapping for observation each type
+    unique_types = df_long["type"].unique()
+    colors = px.colors.qualitative.Plotly
+    color_mapping = {type_: colors[i % len(colors)] for i, type_ in enumerate(unique_types)}
+
+    # Create the figure
+    fig_stat = go.Figure()
+
+    # Loop through each type to add traces
+    for t in df_long["type"].unique():
+        # Filter the DataFrame for this type and stat
+        df_filtered = df_long[(df_long["type"] == t)]
+        
+        # Add a trace
+        fig_stat.add_trace(go.Scatter(
+            x=df_filtered[stat+"_value"],
+            y=df_filtered["midpoint"],
+            mode='lines+markers',
+            name='prior ' + t,
+            line=dict(color=color_mapping[t]),  # Same color for all traces in group
+            marker=dict(size=8, color=color_mapping[t]),
+            legendgroup=t  # Group traces by type
+        ))
+
+    # Update layout
+    fig_stat.update_layout(
+        title=stat + ' by Level',
+        xaxis_title=stat,
+        yaxis_title=verticalUnit,
+        width=800,
+        height=800,
+        template="plotly_white"
+    )
+
+    if verticalUnit == "pressure (Pa)":
+        fig_stat.update_yaxes(autorange="reversed")
+        
+    return fig_stat
+
     
 def mean_then_sqrt(x):
     """
@@ -130,12 +284,36 @@ def mean_then_sqrt(x):
         
     return np.sqrt(np.mean(x))
 
-def rmse_bias(df):
-    rmse_bias_df = df.groupby(['hPa', 'type']).agg({'sq_err':mean_then_sqrt, 'bias':'mean'}).reset_index()
-    rmse_bias_df.rename(columns={'sq_err':'rmse'}, inplace=True)
-    
-    return rmse_bias_df
+def rmse_bias_totalspread(df, phase='prior'): 
+    if phase == 'prior':
+        sq_err_column = 'prior_sq_err'
+        bias_column = 'prior_bias'
+        rmse_column = 'prior_rmse'
+        spread_column = 'prior_ensemble_spread'
+        totalspread_column = 'prior_totalspread'
+    elif phase == 'posterior':
+        sq_err_column = 'posterior_sq_err'
+        bias_column = 'posterior_bias'
+        rmse_column = 'posterior_rmse'
+        spread_column = 'posterior_ensemble_spread'
+        totalspread_column = 'posterior_totalspread'
+    else:
+        raise ValueError("Invalid phase. Must be 'prior' or 'posterior'.")
+
+    rmse_bias_ts_df = df.groupby(['midpoint', 'type'], observed=False).agg({
+        sq_err_column: mean_then_sqrt,
+        bias_column: 'mean',
+        spread_column: mean_then_sqrt,
+        'obs_err_var': mean_then_sqrt     
+    }).reset_index()
 
+    # Add column for totalspread
+    rmse_bias_ts_df[totalspread_column] = np.sqrt(rmse_bias_ts_df[spread_column] + rmse_bias_ts_df['obs_err_var'])
+
+    # Rename square error to root mean square error
+    rmse_bias_ts_df.rename(columns={sq_err_column: rmse_column}, inplace=True)
+    
+    return rmse_bias_ts_df
 
 def rmse_bias_by_obs_type(df, obs_type):
     """
@@ -155,7 +333,7 @@ def rmse_bias_by_obs_type(df, obs_type):
         raise ValueError(f"Observation type '{obs_type}' not found in DataFrame.")
     else:
         obs_type_df = df[df['type'] == obs_type]
-        obs_type_agg = obs_type_df.groupby('plevels').agg({'sq_err':mean_then_sqrt, 'bias':'mean'}).reset_index()
+        obs_type_agg = obs_type_df.groupby('vlevels', observed=False).agg({'sq_err':mean_then_sqrt, 'bias':'mean'}).reset_index()
         obs_type_agg.rename(columns={'sq_err':'rmse'}, inplace=True)
         return obs_type_agg
 

pydartdiags-0.0.43.dist-info/METADATA

@@ -0,0 +1,45 @@
+Metadata-Version: 2.1
+Name: pydartdiags
+Version: 0.0.43
+Summary: Observation Sequence Diagnostics for DART
+Home-page: https://github.com/NCAR/pyDARTdiags.git
+Author: Helen Kershaw
+Author-email: Helen Kershaw <hkershaw@ucar.edu>
+Project-URL: Homepage, https://github.com/NCAR/pyDARTdiags.git
+Project-URL: Issues, https://github.com/NCAR/pyDARTdiags/issues
+Project-URL: Documentation, https://ncar.github.io/pyDARTdiags
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: pandas>=2.2.0
+Requires-Dist: numpy>=1.26
+Requires-Dist: plotly>=5.22.0
+Requires-Dist: pyyaml>=6.0.2
+
+[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
+[![codecov](https://codecov.io/gh/NCAR/pyDARTdiags/graph/badge.svg?token=VK55SQZSVD)](https://codecov.io/gh/NCAR/pyDARTdiags)
+[![PyPI version](https://badge.fury.io/py/pydartdiags.svg)](https://pypi.org/project/pydartdiags/)
+
+
+# pyDARTdiags
+
+pyDARTdiags is a Python library for obsevation space diagnostics for the Data Assimilation Research Testbed ([DART](https://github.com/NCAR/DART)).
+
+pyDARTdiags is under initial development, so please use caution.
+The MATLAB [observation space diagnostics](https://docs.dart.ucar.edu/en/latest/guide/matlab-observation-space.html) are available through [DART](https://github.com/NCAR/DART).
+
+
+pyDARTdiags can be installed through pip: https://pypi.org/project/pydartdiags/  
+Documenation : https://ncar.github.io/pyDARTdiags/
+
+## Contributing
+Contributions are welcome! If you have a feature request, bug report, or a suggestion, please open an issue on our GitHub repository.
+Please read our [Contributors Guide](https://github.com/NCAR/pyDARTdiags/blob/main/CONTRIBUTING.md) if you would like to contribute to
+pyDARTdiags.
+
+## License
+
+pyDARTdiags is released under the Apache License 2.0. For more details, see the LICENSE file in the root directory of this source tree or visit [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0).

pydartdiags-0.0.43.dist-info/RECORD

@@ -0,0 +1,10 @@
+pydartdiags/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+pydartdiags/obs_sequence/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+pydartdiags/obs_sequence/obs_sequence.py,sha256=2pddiJ6VRFkaDizYq8HvGUpC4rw7TTV14XjmemjqCNg,34187
+pydartdiags/plots/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+pydartdiags/plots/plots.py,sha256=UecLgWauO9L_EaGhEVxW3IuKcSU95uRA2mptsxh4-0E,13901
+pydartdiags-0.0.43.dist-info/LICENSE,sha256=ROglds_Eg_ylXp-1MHmEawDqMw_UsCB4r9sk7z9PU9M,11377
+pydartdiags-0.0.43.dist-info/METADATA,sha256=udwmddMTrqFpyj0tjOffWVf2xbTI_3IwQCS4ZVvnnuU,2185
+pydartdiags-0.0.43.dist-info/WHEEL,sha256=A3WOREP4zgxI0fKrHUG8DC8013e3dK3n7a6HDbcEIwE,91
+pydartdiags-0.0.43.dist-info/top_level.txt,sha256=LfMoPLnSd0VhhlWev1eeX9t6AzvyASOloag0LO_ppWg,12
+pydartdiags-0.0.43.dist-info/RECORD,,

{pydartdiags-0.0.42.dist-info → pydartdiags-0.0.43.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.1.0)
+Generator: setuptools (75.7.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

pydartdiags-0.0.42.dist-info/METADATA

@@ -1,404 +0,0 @@
-Metadata-Version: 2.1
-Name: pydartdiags
-Version: 0.0.42
-Summary: Observation Sequence Diagnostics for DART
-Home-page: https://github.com/NCAR/pyDARTdiags.git
-Author: Helen Kershaw
-Author-email: Helen Kershaw <hkershaw@ucar.edu>
-Project-URL: Homepage, https://github.com/NCAR/pyDARTdiags.git
-Project-URL: Issues, https://github.com/NCAR/pyDARTdiags/issues
-Project-URL: Documentation, https://ncar.github.io/pyDARTdiags
-Classifier: Programming Language :: Python :: 3
-Classifier: License :: OSI Approved :: Apache Software License
-Classifier: Operating System :: OS Independent
-Requires-Python: >=3.8
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: pandas>=2.2.0
-Requires-Dist: numpy>=1.26
-Requires-Dist: plotly>=5.22.0
-Requires-Dist: pyyaml>=6.0.2
-
-[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
-[![codecov](https://codecov.io/gh/NCAR/pyDARTdiags/graph/badge.svg?token=VK55SQZSVD)](https://codecov.io/gh/NCAR/pyDARTdiags)
-[![PyPI version](https://badge.fury.io/py/pydartdiags.svg)](https://pypi.org/project/pydartdiags/)
-
-
-# pyDARTdiags
-
-pyDARTdiags is a Python library for obsevation space diagnostics for the Data Assimilation Research Testbed ([DART](https://github.com/NCAR/DART)).
-
-pyDARTdiags is under initial development, so please use caution.
-The MATLAB [observation space diagnostics](https://docs.dart.ucar.edu/en/latest/guide/matlab-observation-space.html) are available through [DART](https://github.com/NCAR/DART).
-
-
-pyDARTdiags can be installed through pip: https://pypi.org/project/pydartdiags/  
-Documenation : https://ncar.github.io/pyDARTdiags/
-
-We recommend installing pydartdiags in a virtual enviroment:
-
-
-```
-python3 -m venv dartdiags
-source dartdiags/bin/activate
-pip install pydartdiags
-```
-
-## Example importing the obs\_sequence and plots modules
-
-```python
-from pydartdiags.obs_sequence import obs_sequence as obsq
-from pydartdiags.plots import plots
-```
-
-## Examining the dataframe
-
-```python
-obs_seq = obsq.obs_sequence('obs_seq.final.ascii')
-obs_seq.df.head()
-```
-
-<table border="1" class="dataframe">
-  <thead>
-    <tr style="text-align: right;">
-      <th></th>
-      <th>obs_num</th>
-      <th>observation</th>
-      <th>prior_ensemble_mean</th>
-      <th>prior_ensemble_spread</th>
-      <th>prior_ensemble_member_1</th>
-      <th>prior_ensemble_member_2</th>
-      <th>prior_ensemble_member_3</th>
-      <th>prior_ensemble_member_4</th>
-      <th>prior_ensemble_member_5</th>
-      <th>prior_ensemble_member_6</th>
-      <th>...</th>
-      <th>latitude</th>
-      <th>vertical</th>
-      <th>vert_unit</th>
-      <th>type</th>
-      <th>seconds</th>
-      <th>days</th>
-      <th>time</th>
-      <th>obs_err_var</th>
-      <th>bias</th>
-      <th>sq_err</th>
-    </tr>
-  </thead>
-  <tbody>
-    <tr>
-      <th>0</th>
-      <td>1</td>
-      <td>230.16</td>
-      <td>231.310652</td>
-      <td>0.405191</td>
-      <td>231.304725</td>
-      <td>231.562874</td>
-      <td>231.333915</td>
-      <td>231.297690</td>
-      <td>232.081416</td>
-      <td>231.051063</td>
-      <td>...</td>
-      <td>0.012188</td>
-      <td>23950.0</td>
-      <td>pressure (Pa)</td>
-      <td>ACARS_TEMPERATURE</td>
-      <td>75603</td>
-      <td>153005</td>
-      <td>2019-12-01 21:00:03</td>
-      <td>1.00</td>
-      <td>1.150652</td>
-      <td>1.324001</td>
-    </tr>
-    <tr>
-      <th>1</th>
-      <td>2</td>
-      <td>18.40</td>
-      <td>15.720527</td>
-      <td>0.630827</td>
-      <td>14.217207</td>
-      <td>15.558196</td>
-      <td>15.805599</td>
-      <td>16.594644</td>
-      <td>14.877743</td>
-      <td>16.334438</td>
-      <td>...</td>
-      <td>0.012188</td>
-      <td>23950.0</td>
-      <td>pressure (Pa)</td>
-      <td>ACARS_U_WIND_COMPONENT</td>
-      <td>75603</td>
-      <td>153005</td>
-      <td>2019-12-01 21:00:03</td>
-      <td>6.25</td>
-      <td>-2.679473</td>
-      <td>7.179578</td>
-    </tr>
-    <tr>
-      <th>2</th>
-      <td>3</td>
-      <td>1.60</td>
-      <td>-4.932073</td>
-      <td>0.825899</td>
-      <td>-5.270562</td>
-      <td>-5.955998</td>
-      <td>-4.209766</td>
-      <td>-5.105016</td>
-      <td>-4.669405</td>
-      <td>-4.365305</td>
-      <td>...</td>
-      <td>0.012188</td>
-      <td>23950.0</td>
-      <td>pressure (Pa)</td>
-      <td>ACARS_V_WIND_COMPONENT</td>
-      <td>75603</td>
-      <td>153005</td>
-      <td>2019-12-01 21:00:03</td>
-      <td>6.25</td>
-      <td>-6.532073</td>
-      <td>42.667980</td>
-    </tr>
-    <tr>
-      <th>3</th>
-      <td>4</td>
-      <td>264.16</td>
-      <td>264.060532</td>
-      <td>0.035584</td>
-      <td>264.107192</td>
-      <td>264.097270</td>
-      <td>264.073212</td>
-      <td>264.047718</td>
-      <td>264.074140</td>
-      <td>264.019895</td>
-      <td>...</td>
-      <td>0.010389</td>
-      <td>56260.0</td>
-      <td>pressure (Pa)</td>
-      <td>ACARS_TEMPERATURE</td>
-      <td>75603</td>
-      <td>153005</td>
-      <td>2019-12-01 21:00:03</td>
-      <td>1.00</td>
-      <td>-0.099468</td>
-      <td>0.009894</td>
-    </tr>
-    <tr>
-      <th>4</th>
-      <td>5</td>
-      <td>11.60</td>
-      <td>10.134115</td>
-      <td>0.063183</td>
-      <td>10.067956</td>
-      <td>10.078798</td>
-      <td>10.120263</td>
-      <td>10.084885</td>
-      <td>10.135112</td>
-      <td>10.140610</td>
-      <td>...</td>
-      <td>0.010389</td>
-      <td>56260.0</td>
-      <td>pressure (Pa)</td>
-      <td>ACARS_U_WIND_COMPONENT</td>
-      <td>75603</td>
-      <td>153005</td>
-      <td>2019-12-01 21:00:03</td>
-      <td>6.25</td>
-      <td>-1.465885</td>
-      <td>2.148818</td>
-    </tr>
-  </tbody>
-</table>
-<p>5 rows × 97 columns</p>
-</div>
-
-
-Find the numeber of assimilated (used) observations vs. possible observations by type
-
-```python
-obsq.possible_vs_used(obs_seq.df)
-```
-
-<table border="1" class="dataframe">
-  <thead>
-    <tr style="text-align: right;">
-      <th></th>
-      <th>type</th>
-      <th>possible</th>
-      <th>used</th>
-    </tr>
-  </thead>
-  <tbody>
-    <tr>
-      <th>0</th>
-      <td>ACARS_TEMPERATURE</td>
-      <td>175429</td>
-      <td>128040</td>
-    </tr>
-    <tr>
-      <th>1</th>
-      <td>ACARS_U_WIND_COMPONENT</td>
-      <td>176120</td>
-      <td>126946</td>
-    </tr>
-    <tr>
-      <th>2</th>
-      <td>ACARS_V_WIND_COMPONENT</td>
-      <td>176120</td>
-      <td>127834</td>
-    </tr>
-    <tr>
-      <th>3</th>
-      <td>AIRCRAFT_TEMPERATURE</td>
-      <td>21335</td>
-      <td>13663</td>
-    </tr>
-    <tr>
-      <th>4</th>
-      <td>AIRCRAFT_U_WIND_COMPONENT</td>
-      <td>21044</td>
-      <td>13694</td>
-    </tr>
-    <tr>
-      <th>5</th>
-      <td>AIRCRAFT_V_WIND_COMPONENT</td>
-      <td>21044</td>
-      <td>13642</td>
-    </tr>
-    <tr>
-      <th>6</th>
-      <td>AIRS_SPECIFIC_HUMIDITY</td>
-      <td>6781</td>
-      <td>0</td>
-    </tr>
-    <tr>
-      <th>7</th>
-      <td>AIRS_TEMPERATURE</td>
-      <td>19583</td>
-      <td>7901</td>
-    </tr>
-    <tr>
-      <th>8</th>
-      <td>GPSRO_REFRACTIVITY</td>
-      <td>81404</td>
-      <td>54626</td>
-    </tr>
-    <tr>
-      <th>9</th>
-      <td>LAND_SFC_ALTIMETER</td>
-      <td>21922</td>
-      <td>0</td>
-    </tr>
-    <tr>
-      <th>10</th>
-      <td>MARINE_SFC_ALTIMETER</td>
-      <td>9987</td>
-      <td>0</td>
-    </tr>
-    <tr>
-      <th>11</th>
-      <td>MARINE_SFC_SPECIFIC_HUMIDITY</td>
-      <td>4196</td>
-      <td>0</td>
-    </tr>
-    <tr>
-      <th>12</th>
-      <td>MARINE_SFC_TEMPERATURE</td>
-      <td>8646</td>
-      <td>0</td>
-    </tr>
-    <tr>
-      <th>13</th>
-      <td>MARINE_SFC_U_WIND_COMPONENT</td>
-      <td>8207</td>
-      <td>0</td>
-    </tr>
-    <tr>
-      <th>14</th>
-      <td>MARINE_SFC_V_WIND_COMPONENT</td>
-      <td>8207</td>
-      <td>0</td>
-    </tr>
-    <tr>
-      <th>15</th>
-      <td>RADIOSONDE_SPECIFIC_HUMIDITY</td>
-      <td>14272</td>
-      <td>0</td>
-    </tr>
-    <tr>
-      <th>16</th>
-      <td>RADIOSONDE_SURFACE_ALTIMETER</td>
-      <td>601</td>
-      <td>0</td>
-    </tr>
-    <tr>
-      <th>17</th>
-      <td>RADIOSONDE_TEMPERATURE</td>
-      <td>29275</td>
-      <td>22228</td>
-    </tr>
-    <tr>
-      <th>18</th>
-      <td>RADIOSONDE_U_WIND_COMPONENT</td>
-      <td>36214</td>
-      <td>27832</td>
-    </tr>
-    <tr>
-      <th>19</th>
-      <td>RADIOSONDE_V_WIND_COMPONENT</td>
-      <td>36214</td>
-      <td>27975</td>
-    </tr>
-    <tr>
-      <th>20</th>
-      <td>SAT_U_WIND_COMPONENT</td>
-      <td>107212</td>
-      <td>82507</td>
-    </tr>
-    <tr>
-      <th>21</th>
-      <td>SAT_V_WIND_COMPONENT</td>
-      <td>107212</td>
-      <td>82647</td>
-    </tr>
-  </tbody>
-</table>
-
-
-## Example plotting
-
-### rank histogram
-
-* Select only observations that were assimliated (QC === 0).
-* plot the rank histogram
-
-```python
-df_qc0 = obsq.select_by_dart_qc(obs_seq.df, 0) 
-plots.plot_rank_histogram(df_qc0)
-```
-![Rank Histogram](https://raw.githubusercontent.com/NCAR/pydartdiags/main/docs/images/rankhist.png)
-
-
-### plot profile of RMSE and Bias 
-
-* Chose levels
-* Select only observations that were assimliated (QC === 0).
-* plot the profiles
-
-```python
-hPalevels = [0.0, 100.0,  150.0, 200.0, 250.0, 300.0, 400.0, 500.0, 700, 850, 925, 1000]# float("inf")] # Pa?
-plevels = [i * 100 for i in hPalevels]
-
-df_qc0 = obsq.select_by_dart_qc(obs_seq.df, 0)  # only qc 0
-df_profile, figrmse, figbias = plots.plot_profile(df_qc0, plevels)
-```
-
-![RMSE Plot](https://raw.githubusercontent.com/NCAR/pydartdiags/main/docs/images/rmse.png)
-
-![Bias Plot](https://raw.githubusercontent.com/NCAR/pydartdiags/main/docs/images/bias.png)
-
-## Contributing
-Contributions are welcome! If you have a feature request, bug report, or a suggestion, please open an issue on our GitHub repository.
-
-## License
-
-pyDARTdiags is released under the Apache License 2.0. For more details, see the LICENSE file in the root directory of this source tree or visit [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0).

pydartdiags-0.0.42.dist-info/RECORD

@@ -1,10 +0,0 @@
-pydartdiags/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-pydartdiags/obs_sequence/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-pydartdiags/obs_sequence/obs_sequence.py,sha256=UBaNMJR3MOro47KyJMdgJ0p-aEqcrp817-SOtpFQ1bg,31511
-pydartdiags/plots/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-pydartdiags/plots/plots.py,sha256=_vZFgQ9qrmtwE_HAP6_nx3pV4JHRdnYckZ5xUxUH4io,6753
-pydartdiags-0.0.42.dist-info/LICENSE,sha256=ROglds_Eg_ylXp-1MHmEawDqMw_UsCB4r9sk7z9PU9M,11377
-pydartdiags-0.0.42.dist-info/METADATA,sha256=oJe2bv1cjoIhlpJQF2tLIIGhS5zu5qJJzFYEZw9ffgQ,9857
-pydartdiags-0.0.42.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
-pydartdiags-0.0.42.dist-info/top_level.txt,sha256=LfMoPLnSd0VhhlWev1eeX9t6AzvyASOloag0LO_ppWg,12
-pydartdiags-0.0.42.dist-info/RECORD,,