Python pandas.notnull() Examples

def _create_observation_variable(individual_selections, choices, partsworth):
    """
    This function handles creating the PyMC3 observation variables.  It also gracefully handles missing observations in individual selections.

    `individual_selections` is a Series of the individuals selections made, starting from 0. It can contain NaNs which represent answer was not provided.

    `choices` is a DataFrame with a hierarchical index: level=0 enumerates the choices, and level=1 displays the profile at a specific choice.
    It's size is (n_questions, n_choices_per_question).

    `partsworth` is a slice of PyMC3 matrix. It represents the partsworth variables of a individual. Size is (n_profiles,)

    This computes the values exp(partsworth * profile_j) / sum[ exp(partsworth * profile_k ] for all j.
    """
    nan_mask = pd.notnull(individual_selections)
    return pm.Categorical("Obs_%s" % individual_selections.name,
                          tt.nnet.softmax(tt.stack([
                            tt.dot(choice.values, partsworth) for _, choice in choices[nan_mask.values].groupby(axis=1, level=0)
                          ], axis=0).T),
                          observed=individual_selections[nan_mask.values].values) 

def load_geodataframe(geo_filename):
	""" 
	Load input GeoDataFrame

	Parameters
	----------
	geo_filename : string
		input GeoDataFrame filename

	Returns
	----------
	geopandas.GeoDataFrame
		loaded data

	"""
	# Load using geopandas
	df_osm_data = gpd.read_file(geo_filename)
	# Set None as NaN
	df_osm_data.fillna(value=np.nan, inplace=True)
	# Replace empty string (Json NULL sometimes read as '') for NaN
	df_osm_data.replace('', np.nan, inplace=True)
	
	def list_int_from_string(x): # List of integers given input in string format
		return [ int(id_) for id_ in x.split(",") ]
	def list_str_from_string(x): # List of strings given input in string format
		return x.split(",")

	# Recover list
	if ( "activity_category" in df_osm_data.columns): 
		df_osm_data[ "activity_category" ] = df_osm_data.activity_category.apply(lambda x: list_str_from_string(x) if pd.notnull(x) else np.nan )
	if ( "containing_parts" in df_osm_data.columns): 
		df_osm_data[ "containing_parts" ] = df_osm_data.containing_parts.apply( lambda x: list_int_from_string(x) if pd.notnull(x) else np.nan )
	if ( "containing_poi" in df_osm_data.columns): 
		df_osm_data[ "containing_poi" ] = df_osm_data.containing_poi.apply( lambda x: list_int_from_string(x) if pd.notnull(x) else np.nan )
	
	# To UTM coordinates
	return ox.project_gdf( df_osm_data ) 

def fillna(series_or_arr, missing_value=0.0):
    """Fill missing values in pandas objects and numpy arrays.

    Arguments
    ---------
    series_or_arr : pandas.Series, numpy.ndarray
        The numpy array or pandas series for which the missing values
        need to be replaced.
    missing_value : float, int, str
        The value to replace the missing value with. Default 0.0.

    Returns
    -------
    pandas.Series, numpy.ndarray
        The numpy array or pandas series with the missing values
        filled.
    """

    if pandas.notnull(missing_value):
        if isinstance(series_or_arr, (numpy.ndarray)):
            series_or_arr[numpy.isnan(series_or_arr)] = missing_value
        else:
            series_or_arr.fillna(missing_value, inplace=True)

    return series_or_arr 

def load_metadata(self):
    try:
      tasks_filename, metadata_filename = self._get_metadata_filename()
      with open(tasks_filename) as fin:
        tasks = json.load(fin)
      metadata_df = pd.read_csv(metadata_filename, compression='gzip')
      metadata_df = metadata_df.where((pd.notnull(metadata_df)), None)
      return tasks, metadata_df
    except Exception as e:
      pass

    # Load obsolete format -> save in new format
    metadata_filename = os.path.join(self.data_dir, "metadata.joblib")
    if os.path.exists(metadata_filename):
      tasks, metadata_df = load_from_disk(metadata_filename)
      del metadata_df['task_names']
      del metadata_df['basename']
      save_metadata(tasks, metadata_df, self.data_dir)
      return tasks, metadata_df
    raise ValueError("No Metadata Found On Disk") 

def _make_json_dataset(self, edges, nodes, name):
        (elist, nlist) = self._bind_attributes_v1(edges, nodes)
        edict = elist.where((pandas.notnull(elist)), None).to_dict(orient='records')

        bindings = {'idField': self._node or Plotter._defaultNodeId,
                    'destinationField': self._destination, 'sourceField': self._source}
        dataset = {'name': PyGraphistry._config['dataset_prefix'] + name,
                   'bindings': bindings, 'type': 'edgelist', 'graph': edict}

        if nlist is not None:
            ndict = nlist.where((pandas.notnull(nlist)), None).to_dict(orient='records')
            dataset['labels'] = ndict
        return dataset


    # Main helper for creating ETL2 payload 

def objectEncoder(vg, series, dtype):
    series.where(pandas.notnull(series), '\0', inplace=True)
    # vec is a string[] submessage within a repeated
    vec = vg.string_vectors.add()
    str_series = None    
    try:
        str_series = series.astype('unicode')
    except UnicodeDecodeError:
        warnings.warn("Warning: escaping unicode")
        str_series = series.apply(lambda v: v.decode('utf-8'))
    for val in str_series:
        vec.values.append(val)
    return (vec, {'ctype': 'utf8'})


# NaN (as well as Infinity and undefined) are valid JSON. Use this guard to filter
# them out when creating the json metadata. 

def load_metadata(self):
    try:
      tasks_filename, metadata_filename = self._get_metadata_filename()
      with open(tasks_filename) as fin:
        tasks = json.load(fin)
      metadata_df = pd.read_csv(metadata_filename, compression='gzip')
      metadata_df = metadata_df.where((pd.notnull(metadata_df)), None)
      return tasks, metadata_df
    except Exception as e:
      pass

    # Load obsolete format -> save in new format
    metadata_filename = os.path.join(self.data_dir, "metadata.joblib")
    if os.path.exists(metadata_filename):
      tasks, metadata_df = load_from_disk(metadata_filename)
      del metadata_df['task_names']
      del metadata_df['basename']
      save_metadata(tasks, metadata_df, self.data_dir)
      return tasks, metadata_df
    raise ValueError("No Metadata Found On Disk") 

def test_pertmeth(self):
        # Test with specified perturbation method.

        df = gendat()
        orig = df.copy()
        mx = pd.notnull(df)
        nrow, ncol = df.shape

        for pert_meth in "gaussian", "boot":

            imp_data = mice.MICEData(df, perturbation_method=pert_meth)

            for k in range(2):
                imp_data.update_all()
                assert_equal(imp_data.data.shape[0], nrow)
                assert_equal(imp_data.data.shape[1], ncol)
                assert_allclose(orig[mx], imp_data.data[mx])

        assert_equal(imp_data._cycle_order, ['x5', 'x3', 'x4', 'y', 'x2', 'x1']) 

def uniprot_reviewed_checker(uniprot_id):
    """Check if a single UniProt ID is reviewed or not.

    Args:
        uniprot_id:

    Returns:
        bool: If the entry is reviewed

    """

    query_string = 'id:' + uniprot_id

    uni_rev_raw = StringIO(bsup.search(query_string, columns='id,reviewed', frmt='tab'))
    uni_rev_df = pd.read_table(uni_rev_raw, sep='\t', index_col=0)
    uni_rev_df = uni_rev_df.fillna(False)
    uni_rev_df = uni_rev_df[pd.notnull(uni_rev_df.Status)]

    uni_rev_df = uni_rev_df.replace(to_replace="reviewed", value=True)
    uni_rev_df = uni_rev_df.replace(to_replace="unreviewed", value=False)
    uni_rev_dict_adder = uni_rev_df.to_dict()['Status']

    return uni_rev_dict_adder[uniprot_id] 

def parse_psqs(psqs_results_file):
    """Parse a PSQS result file and returns a Pandas DataFrame of the results

    Args:
        psqs_results_file: Path to psqs results file

    Returns:
        Pandas DataFrame: Summary of PSQS results

    """

    # TODO: generalize column names for all results, save as dict instead

    psqs_results = pd.read_csv(psqs_results_file, sep='\t', header=None)
    psqs_results['pdb_file'] = psqs_results[0].apply(lambda x: str(x).strip('./').strip('.pdb'))
    psqs_results = psqs_results.rename(columns = {1:'psqs_local', 2:'psqs_burial', 3:'psqs_contact', 4:'psqs_total'}).drop(0, axis=1)
    psqs_results['u_pdb'] = psqs_results['pdb_file'].apply(lambda x: x.upper() if len(x)==4 else np.nan)
    psqs_results['i_entry_name'] = psqs_results['pdb_file'].apply(lambda x: x.split('_model1')[0] if len(x)>4 else np.nan)
    psqs_results = psqs_results[pd.notnull(psqs_results.psqs_total)]

    return psqs_results 

def execute_internal(self, context, **kwargs):
        """
        the internal execution process to be implemented
        :param context:
        :param kwargs:
        :return:
        """
        df = pd.read_csv('https://raw.githubusercontent.com/bailaohe/parade/master/assets/movie_metadata.csv')

        # Process projection on the dataset to get our interested attributes
        df = df[['movie_title', 'genres', 'title_year', 'content_rating', 'budget', 'num_voted_users', 'imdb_score']]

        # Filter out records with *NAN* title_year and budget
        df = df[pd.notnull(df['title_year'])]
        df = df[df['budget'] > 0]

        # Extract the genres ROOT
        df['genres_root'] = df['genres'].apply(lambda g: g.split('|')[0])

        return df 

def test_df_equivalent_after_sql(self):

        # Parse the CSV
        df_source = services.load_df_from_csvfile(
            io.StringIO(self.csv1),
            0,
            0)

        # Store the DF in the DB
        pandas.store_table(df_source, self.table_name)

        # Load it from the DB
        df_dst = pandas.load_table(self.table_name)

        # NaN in boolean columns are now None
        df_source['bool1'] = df_source['bool1'].where(
            pd.notnull(df_source['bool1']),
            None)
        df_source['bool2'] = df_source['bool2'].where(
            pd.notnull(df_source['bool2']),
            None)

        # Data frames mut be identical
        assert df_source.equals(df_dst) 

def clean_and_write_dataframe_to_csv(data, filename):
    """
    Cleans a dataframe of np.NaNs and saves to file via pandas.to_csv

    :param data: data to write to CSV
    :type data: :class:`pandas.DataFrame`
    :param filename: Path to file to write CSV to. if None, string of data
        will be returned
    :type filename: str | None
    :return: If the filename is None, returns the string of data. Otherwise
        returns None.
    :rtype: str | None
    """
    # cleans np.NaN values
    data = data.where((pd.notnull(data)), None)
    # If filename=None, to_csv will return a string
    result = data.to_csv(path_or_buf=filename, encoding='utf-8', dtype=str, index=False, na_rep=None,
                         skipinitialspace=True, quoting=csv.QUOTE_ALL)
    logging.info("Dataframe of shape %s has been stored." % str(data.shape))

    return result 

def pre_processing(self, labels):
        # removing NaN in lat and lon
        self.raw_data = self.raw_data.loc[pd.notnull(self.raw_data.lat), :]
        self.raw_data = self.raw_data.loc[pd.notnull(self.raw_data.lon), :]
        for label in labels:
            self.raw_data = self.raw_data.loc[pd.notnull(self.raw_data[label]), :]
        """
        lat_= self.raw_data.lat.rolling(3, min_periods=1).median()
        self.raw_data.assign(lat=lat_)
        lon_ = self.raw_data.lon.rolling(3, min_periods=1).median()
        self.raw_data.assign(lot=lon_)

        self.raw_data = self.raw_data.loc[pd.notnull(self.raw_data.lat), :]
        self.raw_data = self.raw_data.loc[pd.notnull(self.raw_data.lon), :]
        """

        return None 

def generate_avro(src_file: Text, output_file: Text):
  """Generates avro file based on src file.

  Args:
    src_file: path to Chicago taxi dataset.
    output_file: output path for avro file.
  """
  df = pd.read_csv(src_file)
  # Replaces NaN's with None's for avroWriter to interpret null values
  df = df.where((pd.notnull(df)), None)

  records = df.to_dict(orient='records')

  parsed_schema = fastavro.parse_schema(get_schema())
  with open(output_file, 'wb') as f:
    fastavro.writer(f, parsed_schema, records) 

def _make_index_names(self, name1, name2):

        if pandas.notnull(name1) and pandas.notnull(name2) and \
                (name1 == name2):
            return ["{}{}".format(name1, self.suffixes[0]),
                    "{}{}".format(name1, self.suffixes[1])]
        else:
            return [name1, name2] 

def load_captions(self):
        df = pd.read_csv(self.caption_fpath)
        df = df[df['Language'] == 'English']
        df = df[pd.notnull(df['Description'])]
        captions = df['Description'].values
        return captions 

def load_captions(self):
        df = pd.read_csv(self.caption_fpath)
        df = df[df['Language'] == 'English']
        df = df[[ 'VideoID', 'Start', 'End', 'Description' ]]
        df = df[pd.notnull(df['Description'])]

        for video_id, start, end, caption in df.values:
            vid = "{}_{}_{}".format(video_id, start, end)
            self.captions[vid].append(caption) 

def load_metadata():
    df = pd.read_csv(C.caption_fpath)
    df = df[df['Language'] == 'English']
    df = df[pd.notnull(df['Description'])]
    df = df.reset_index(drop=True)
    return df 

def test_default(self):
        # Test with all defaults.

        df = gendat()
        orig = df.copy()
        mx = pd.notnull(df)
        imp_data = mice.MICEData(df)
        nrow, ncol = df.shape

        assert_allclose(imp_data.ix_miss['x1'], np.arange(60))
        assert_allclose(imp_data.ix_obs['x1'], np.arange(60, 200))
        assert_allclose(imp_data.ix_miss['x2'], np.arange(40))
        assert_allclose(imp_data.ix_miss['x3'], np.arange(10, 30, 2))
        assert_allclose(imp_data.ix_obs['x3'],
                        np.concatenate((np.arange(10),
                                        np.arange(11, 30, 2),
                                        np.arange(30, 200))))

        for k in range(3):
            imp_data.update_all()
            assert_equal(imp_data.data.shape[0], nrow)
            assert_equal(imp_data.data.shape[1], ncol)
            assert_allclose(orig[mx], imp_data.data[mx])

        fml = 'x1 ~ x2 + x3 + x4 + x5 + y'
        assert_equal(imp_data.conditional_formula['x1'], fml)

        assert_equal(imp_data._cycle_order, ['x5', 'x3', 'x4', 'y', 'x2', 'x1'])

        # Should make a copy
        assert(not (df is imp_data.data))

        (endog_obs, exog_obs, exog_miss,
         predict_obs_kwds, predict_miss_kwds) = imp_data.get_split_data('x3')
        assert_equal(len(endog_obs), 190)
        assert_equal(exog_obs.shape, [190, 6])
        assert_equal(exog_miss.shape, [10, 6]) 

def test_set_imputer(self):
        # Test with specified perturbation method.

        from statsmodels.regression.linear_model import RegressionResultsWrapper
        from statsmodels.genmod.generalized_linear_model import GLMResultsWrapper

        df = gendat()
        orig = df.copy()
        mx = pd.notnull(df)
        nrow, ncol = df.shape

        imp_data = mice.MICEData(df)
        imp_data.set_imputer('x1', 'x3 + x4 + x3*x4')
        imp_data.set_imputer('x2', 'x4 + I(x5**2)')
        imp_data.set_imputer('x3', model_class=sm.GLM,
                             init_kwds={"family": sm.families.Binomial()})

        imp_data.update_all()
        assert_equal(imp_data.data.shape[0], nrow)
        assert_equal(imp_data.data.shape[1], ncol)
        assert_allclose(orig[mx], imp_data.data[mx])
        for j in range(1, 6):
            if j == 3:
                assert_equal(isinstance(imp_data.models['x3'], sm.GLM), True)
                assert_equal(isinstance(imp_data.models['x3'].family, sm.families.Binomial), True)
                assert_equal(isinstance(imp_data.results['x3'], GLMResultsWrapper), True)
            else:
                assert_equal(isinstance(imp_data.models['x%d' % j], sm.OLS), True)
                assert_equal(isinstance(imp_data.results['x%d' % j], RegressionResultsWrapper), True)

        fml = 'x1 ~ x3 + x4 + x3*x4'
        assert_equal(imp_data.conditional_formula['x1'], fml)

        fml = 'x4 ~ x1 + x2 + x3 + x5 + y'
        assert_equal(imp_data.conditional_formula['x4'], fml)

        assert_equal(imp_data._cycle_order, ['x5', 'x3', 'x4', 'y', 'x2', 'x1']) 

def _handle_missing(data, groups, formula, re_formula, vc_formula):

    tokens = set([])

    forms = [formula]
    if re_formula is not None:
        forms.append(re_formula)
    if vc_formula is not None:
        forms.extend(vc_formula.values())

    import tokenize
    from statsmodels.compat import PY3
    from statsmodels.compat.python import StringIO, asunicode
    skiptoks = {"(", ")", "*", ":", "+", "-", "**", "/"}

    for fml in forms:
        # Unicode conversion is for Py2 compatability
        rl = StringIO(fml)

        def rlu():
            line = rl.readline()
            return asunicode(line, 'ascii')
        g = tokenize.generate_tokens(rlu)
        for tok in g:
            if tok not in skiptoks:
                if PY3:
                    tokens.add(tok.string)
                else:
                    tokens.add(tok[1])
    tokens = list(tokens & set(data.columns))
    tokens.sort()

    data = data[tokens]
    ii = pd.notnull(data).all(1)
    if type(groups) != "str":
        ii &= pd.notnull(groups)

    return data.loc[ii, :], groups[np.asarray(ii)] 

def format_background_colormap(self, col, colormap, vmin, vmax):

        iCol, worksheetCol = self.convert_to_col_index(col)

        for index, row in self.df.iterrows():
            x = row.iloc[iCol]
            if pd.notnull(x):
                colorHex = convert_colormap_to_hex(colormap, x, vmin=vmin, vmax=vmax)
                rowIndex = self.df.index.get_loc(index)
                self.formatTable[rowIndex][iCol]['bg_color'] = colorHex 

def _merge_metrics(row):
  non_empty = [str(r) for r in row if pd.notnull(r)]
  if not non_empty:
    return None
  elif len(non_empty) == 1:
    return non_empty[0]
  else:
    return "::".join(non_empty)


# TODO(dlsun): Remove AnalysisParameters and incorporate the
#   attributes directly into the Analyze object. 

def test_custom_grouper(self):

        dti = DatetimeIndex(freq='Min', start=datetime(2005, 1, 1),
                            end=datetime(2005, 1, 10))

        s = Series(np.array([1] * len(dti)), index=dti, dtype='int64')

        b = TimeGrouper(Minute(5))
        g = s.groupby(b)

        # check all cython functions work
        funcs = ['add', 'mean', 'prod', 'ohlc', 'min', 'max', 'var']
        for f in funcs:
            g._cython_agg_general(f)

        b = TimeGrouper(Minute(5), closed='right', label='right')
        g = s.groupby(b)
        # check all cython functions work
        funcs = ['add', 'mean', 'prod', 'ohlc', 'min', 'max', 'var']
        for f in funcs:
            g._cython_agg_general(f)

        self.assertEquals(g.ngroups, 2593)
        self.assert_(notnull(g.mean()).all())

        # construct expected val
        arr = [1] + [5] * 2592
        idx = dti[0:-1:5]
        idx = idx.append(dti[-1:])
        expect = Series(arr, index=idx)

        # GH2763 - return in put dtype if we can
        result = g.agg(np.sum)
        assert_series_equal(result, expect)

        df = DataFrame(np.random.rand(len(dti), 10), index=dti, dtype='float64')
        r = df.groupby(b).agg(np.sum)

        self.assertEquals(len(r.columns), 10)
        self.assertEquals(len(r.index), 2593) 

def _check_expanding_ndarray(self, func, static_comp, has_min_periods=True,
                                 has_time_rule=True, preserve_nan=True):
        result = func(self.arr)

        assert_almost_equal(result[10],
                            static_comp(self.arr[:11]))

        if preserve_nan:
            assert(np.isnan(result[self._nan_locs]).all())

        arr = randn(50)

        if has_min_periods:
            result = func(arr, min_periods=30)
            assert(np.isnan(result[:29]).all())
            assert_almost_equal(result[-1], static_comp(arr[:50]))

            # min_periods is working correctly
            result = func(arr, min_periods=15)
            self.assert_(np.isnan(result[13]))
            self.assert_(not np.isnan(result[14]))

            arr2 = randn(20)
            result = func(arr2, min_periods=5)
            self.assert_(isnull(result[3]))
            self.assert_(notnull(result[4]))

            # min_periods=0
            result0 = func(arr, min_periods=0)
            result1 = func(arr, min_periods=1)
            assert_almost_equal(result0, result1)
        else:
            result = func(arr)
            assert_almost_equal(result[-1], static_comp(arr[:50])) 

def test_longpanel_series_combo(self):
        wp = tm.makePanel()
        lp = wp.to_frame()

        y = lp.pop('ItemA')
        model = ols(y=y, x=lp, entity_effects=True, window=20)
        self.assert_(notnull(model.beta.values).all())
        tm.assert_isinstance(model, PanelOLS)
        model.summary 

def after_all(self, context_all):
    # add col for utilization in percentage
    def calc_usedPct(row):
      if row.capacity_usd==0: return 0
      o = row.used_usd / row.capacity_usd * 100
      return int(o)

    self.df_bins['used_pct'] = self.df_bins.apply(calc_usedPct, axis=1)

    # add column for regions as string
    self.df_bins['regions_str'] = self.df_bins['regions_set'].apply(lambda x: "0" if len(x)==0 else "%i (%s)"%(len(x), l2s(x)))

    # cases where dt_start > dt_end are those where there was no data and the initialization remained
    # so overwrite with na
    # Update 2019-12-11 Now that the df_bins timestamps are set with resample and dt_end is inclusive,
    # instead of setting to na, just swap the start/end fake timestamps which represent the end/start of the periods
    #import numpy as np
    #self.df_bins['dt_start'] = self.df_bins.apply(lambda row: np.nan if row.count_analyzed==0 else row.dt_start, axis=1)
    #self.df_bins['dt_end']   = self.df_bins.apply(lambda row: np.nan if row.count_analyzed==0 else row.dt_end  , axis=1)
    self.df_bins['dt_start_bkp']  = self.df_bins['dt_start']
    self.df_bins['dt_start'] = self.df_bins.apply(lambda row: row.dt_end       if row.count_analyzed==0 else row.dt_start, axis=1)
    self.df_bins['dt_end']   = self.df_bins.apply(lambda row: row.dt_start_bkp if row.count_analyzed==0 else row.dt_end  , axis=1)
    del self.df_bins['dt_start_bkp']

    # convert the dt_{start,end} back to dates again, given the nans
    for fx in ['dt_start', 'dt_end']: self.df_bins[fx] = pd.to_datetime(self.df_bins[fx])

    # Bugfix for cloudwatch:
    # When 90>=ndays>=64, cloudwatchman/metric.get_statistics returns data with max Timestamp on mainmanager.EndTime
    # Otherwise, the data max Timestamp is EndTime - 1 day
    # Here, get around this problem by incrementing by 1 day (or just set to mainmanager.EndTime)
    # This was tested on 2019-12-10 10:00 am UTC, and the last date was Dec 9 for ndays<64 and Dec 10 for ndays>=64
    if pd.notnull(self.df_bins['dt_end'].iloc[-1]):
      dt_max = context_all['mainManager'].EndTime.date()
      import datetime as dt
      dt_lastp1 = self.df_bins.dt_end.iloc[-1].date() + dt.timedelta(days=1)
      self.df_bins.iloc[-1, self.df_bins.columns=='dt_end'] = min(dt_max, dt_lastp1)

    # inject result for reporter access
    context_all['df_bins'] = self.df_bins
    return context_all 

def import_df(self, df):
        '''Import LabelTree from DataFrame
        
        Args:
            df (pandas.DataFrame): LabelTree in DataFrame style.

        Retruns:
            :class:`lost.db.model.LabelLeaf` or None: 
                The created root leaf or None if a root leaf with same
                name is already present in database.
        '''
        df = df.where((pd.notnull(df)), None)
        root = df[df['parent_leaf_id'].isnull()]
        no_root = df[~df['parent_leaf_id'].isnull()]
        childs = {}

        if len(root) != 1:
            raise ValueError('''Can not import. There needs 
                to be exactly one root leaf for that tree! 
                Found: \n{}'''.format(root))
        else:
            try:
                root_leaf = self.create_root(root['name'].values[0])
                if root_leaf is None:
                    return None #A tree with the same name already exists.
                self._df_row_to_leaf(root.loc[0], root_leaf)

                #Create child dict
                for index, row in no_root.iterrows():
                    if not row['parent_leaf_id'] in childs:
                        childs[row['parent_leaf_id']] = []
                    childs[row['parent_leaf_id']].append(row)
                
                self.__create_childs_from_df(childs, root_leaf, root.loc[0])
                self.dbm.commit()
                return root_leaf
            except KeyError:
                self.logger.error('''At least the following columns 
                    need to be provided: *idx*, *name*, *parent_leaf_id*''')
                raise 

def _split(value, separators):
        values = []
        for separator in separators:
            if separator in str(value):
                values = [str(split_value) for split_value in
                          value.split(separator)]
                break

        return pd.Series([str(value).strip() for value in values
                          if pd.notnull(value)])