Merge pull request #72 from holukas/anomaly-plot

Anomaly plot
holukas · Mar 19, 2024 · 191d318 · 191d318
2 parents ade67d3 + d503b9b
commit 191d318
Show file tree

Hide file tree

Showing 11 changed files with 499 additions and 30 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -2,6 +2,18 @@
 
 ![DIIVE](images/logo_diive1_256px.png)
 
+## v0.71.4 | 20 Mar 2024
+
+### Changes
+
+- Refactored class `LongtermAnomaliesYear` (`diive.core.plotting.bar.LongtermAnomaliesYear`)
+
+![DIIVE](images/plotBarLongtermAnomaliesYear_diive_v0.71.4.png)
+
+### Notebooks
+
+- Added new notebook for `LongtermAnomaliesYear` (`notebooks/Plotting/LongTermAnomalies.ipynb`)
+
 ## v0.71.3 | 19 Mar 2024
 
 ### Changes

diff --git a/README.md b/README.md
@@ -109,6 +109,7 @@ Fill gaps in time series with various methods
 
 ### Plotting
 
+- Long-term anomalies per year ([notebook example](https://github.com/holukas/diive/blob/main/notebooks/Plotting/LongTermAnomalies.ipynb))
 - Simple (interactive) time series
   plot ([notebook example](https://github.com/holukas/diive/blob/main/notebooks/Plotting/TimeSeries.ipynb))
 - ScatterXY plot ([notebook example](https://github.com/holukas/diive/blob/main/notebooks/Plotting/ScatterXY.ipynb))

diff --git a/diive/core/plotting/bar.py b/diive/core/plotting/bar.py
@@ -9,11 +9,25 @@ class LongtermAnomaliesYear:
 
     def __init__(self,
                  series: Series,
-                 series_units: str,
                  reference_start_year: int,
-                 reference_end_year: int):
+                 reference_end_year: int,
+                 series_label: str = None,
+                 series_units: str = None):
+        """Calculate and plot long-term anomaly for a variable, per year, compared to a reference period.
+
+        Args:
+            series: Time series for anomalies with one value per year.
+            reference_start_year: First year of the reference period.
+            reference_end_year: Last year of the reference period.
+            series_label: Label for *series* on the y-axis of the plot.
+            series_units: Units for *series* on the y-axis of the plot.
+
+        - Example notebook available in:
+            notebooks/Plotting/LongTermAnomalies.ipynb
+        """
         self.series = series
         self.series_units = series_units
+        self.series_label = series_label
         self.reference_start_year = reference_start_year
         self.reference_end_year = reference_end_year
 
@@ -27,7 +41,7 @@ def __init__(self,
         self.anomalies_df = self._calc_reference()
 
     def _apply_format(self):
-        title = f"Anomaly per year ({self.data_first_year}-{self.data_last_year})"
+        title = f"{self.series_label} anomaly per year ({self.data_first_year}-{self.data_last_year})"
         self.fig.suptitle(title, fontsize=theme.FIGHEADER_FONTSIZE)
 
         ref_mean = self.anomalies_df['reference_mean'].iloc[-1]
@@ -47,7 +61,7 @@ def _apply_format(self):
         self.ax.locator_params(axis='x', nbins=nbins)
         pf.default_format(ax=self.ax,
                           ax_xlabel_txt='Year',
-                          ax_ylabel_txt=self.series.name,
+                          ax_ylabel_txt=f"{self.series_label} anomaly",
                           txt_ylabel_units=self.series_units,
                           showgrid=False)
         self.ax.axhline(0, lw=1, color='black')
@@ -74,8 +88,10 @@ def get(self):
 
     def plot(self, showplot: bool = True):
         # ax1.plot(ta_longterm.index.values, ta_longterm['diff'].values)
-        self.anomalies_df['anomaly_above'].plot.bar(color='#F44336', ax=self.ax, legend=False)
-        self.anomalies_df['anomaly_below'].plot.bar(color='#2196F3', ax=self.ax, legend=False)
+        self.anomalies_df['anomaly_above'].plot.bar(color='#EF5350', ax=self.ax, legend=False, width=.7)
+        # self.anomalies_df['anomaly_above'].plot.bar(color='#F44336', ax=self.ax, legend=False)
+        self.anomalies_df['anomaly_below'].plot.bar(color='#42A5F5', ax=self.ax, legend=False, width=.7)
+        # self.anomalies_df['anomaly_below'].plot.bar(color='#2196F3', ax=self.ax, legend=False)
         # ta_longterm_anomalies.plot.bar(x='year', y='Temperature', color='#2196F3', ax=ax1)
         # ta_longterm_anomalies_above.plot.bar(x='year', y='Temperature', color='red', ax=ax1)
         # ta_longterm_anomalies_below.plot.bar(x='year', y='Temperature', color='blue', ax=ax1)
@@ -89,15 +105,19 @@ def plot(self, showplot: bool = True):
 def example():
     ## Long-term TA
     ## space-separated data
-    data_longterm_TA = r"F:\Dropbox\luhk_work\_current\CH-DAV_1864-2021_TA-YEARLY_Meteoswiss_order_105469_data.txt"
-    ta_longterm = pd.read_csv(data_longterm_TA, header=0, encoding='utf-8', delimiter=';',
-                              keep_date_col=False, index_col='time', dtype=None,
-                              engine='python')
-    ta_longterm = ta_longterm['tre200y0'].copy()
-    LongtermAnomaliesYear(series=ta_longterm,
+    data_longterm_TA = r"L:\Sync\luhk_work\80 - SITES\CH-DAV\Data\Datasets\MeteoSwiss\CH-DAV_1864-2023_TA-YEARLY_Meteoswiss_order_120443_data.txt"
+    df = pd.read_csv(data_longterm_TA, header=0, encoding='utf-8', delimiter=';',
+                     keep_date_col=False, index_col='time', dtype=None,
+                     engine='python')
+
+    series = df['tre200y0'].copy()
+    series_label = "Air temperature"
+    LongtermAnomaliesYear(series=series,
+                          series_label=series_label,
                           series_units='(°C)',
                           reference_start_year=1864,
                           reference_end_year=1913).plot()
 
+
 if __name__ == '__main__':
     example()
diff --git a/diive/core/plotting/plotfuncs.py b/diive/core/plotting/plotfuncs.py
@@ -509,7 +509,7 @@ def save_fig(fig,
 
 
 def create_ax(facecolor: str = 'white',
-              figsize: tuple = (8, 4.5),
+              figsize: tuple = (14, 8),
               dpi: int = 100):
     """Create figure and axis"""
     # Figure setup

diff --git a/diive/pkgs/analyses/decoupling.py b/diive/pkgs/analyses/decoupling.py
@@ -34,18 +34,21 @@ def __init__(self,
                  n_bins_var1: int = 48,
                  n_subbins_var2: int = 2,
                  convert_to_percentiles: bool = False):
-        """Investigate binned aggregates of a variable z in binned classes of x and y.
+        """Investigate binned aggregates (median) of a variable z in binned classes of x and y.
 
-        For example: show mean GPP (y) in 5 classes of VPD (x), separate for 10 classes
+        For example: show median GPP (y) in 5 classes of VPD (x), separate for 10 classes
         of air temperature (z).
 
         Args:
-            df:
-            var1_col:
-            var2_col:
-            var3_col:
-            n_bins_var1:
-            n_subbins_var2:
+            df: Dataframe with variables
+            var1_col: Name of the first binning variable, will be shown as colors (z) in the plot.
+            var2_col: Name of the second binning variable, will be shown on the x-axis (x) in the plot.
+            var3_col: Name of the variable of interest, will be shown on the y-axis (y) in the plot.
+            n_bins_var1: Number of bins for variable *var1_col*.
+            n_subbins_var2: Number of bins for variable *var2_col*.
+
+        Returns:
+            Dict with results stored as dataframes for each bin of *var1_col*.
 
         - Example notebook available in:
             notebooks/Analyses/DecouplingSortingBins.ipynb
@@ -78,7 +81,7 @@ def binmedians(self) -> dict:
             raise Exception('No binned means available, try to run .calcbins() first.')
         return self._binmedians
 
-    def get_binmeans(self) -> dict:
+    def get_binmedians(self) -> dict:
         """Return dict of dataframes with variable 1 group as key"""
         return self.binmedians
 
@@ -269,7 +272,7 @@ def example():
     sbm.calcbins()
     sbm.showplot_decoupling_sbm(marker='o', emphasize_lines=True)
 
-    binmedians = sbm.get_binmeans()
+    binmedians = sbm.get_binmedians()
     first = next(iter(binmedians))
     print(binmedians[first])
 

diff --git a/images/plotBarLongtermAnomaliesYear_diive_v0.71.4.png b/images/plotBarLongtermAnomaliesYear_diive_v0.71.4.png
diff --git a/notebooks/Analyses/DecouplingSortingBins.ipynb b/notebooks/Analyses/DecouplingSortingBins.ipynb
@@ -80,7 +80,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "diive version: v0.71.3\n"
+      "diive version: v0.71.4\n"
      ]
     }
    ],
@@ -137,7 +137,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Loaded .parquet file L:\\Sync\\luhk_work\\20 - CODING\\21 - DIIVE\\diive\\diive\\configs\\exampledata\\exampledata_CH-DAV_FP2022.5_2013-2022_ID20230206154316_30MIN.parquet (0.055 seconds). Detected time resolution of <30 * Minutes> / 30T \n"
+      "Loaded .parquet file L:\\Sync\\luhk_work\\20 - CODING\\21 - DIIVE\\diive\\diive\\configs\\exampledata\\exampledata_CH-DAV_FP2022.5_2013-2022_ID20230206154316_30MIN.parquet (0.092 seconds). Detected time resolution of <30 * Minutes> / 30T \n"
      ]
     },
     {
@@ -313,7 +313,7 @@
    "id": "df38fda9-066a-4fef-986a-cfe921fb88c0",
    "metadata": {},
    "source": [
-    "# (1) **Calculate daily correlation between `Rg_f` and `SW_IN_POT`**"
+    "# **Calculate bin medians**"
    ]
   },
   {
@@ -369,7 +369,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "binmedians = sbm.get_binmeans()"
+    "binmedians = sbm.get_binmedians()"
    ]
   },
   {
@@ -530,7 +530,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Finished 2024-03-19 17:06:54\n"
+      "Finished 2024-03-19 23:56:08\n"
      ]
     }
    ],

diff --git a/notebooks/OVERVIEW.ipynb b/notebooks/OVERVIEW.ipynb
@@ -26,7 +26,7 @@
    "metadata": {},
    "source": [
     "---\n",
-    "**Last updated:**: 19 Mar 2024  \n",
+    "**Last updated:**: 20 Mar 2024  \n",
     "**Author**: Lukas Hörtnagl (holukas@ethz.ch)  \n",
     "Overview of example notebooks for the time series processing library `diive`."
    ]
@@ -237,6 +237,7 @@
    "id": "3f12e3ac-15eb-4dcb-bbbc-4652149b25d9",
    "metadata": {},
    "source": [
+    "- [Long-term anomalies per year](Plotting/LongTermAnomalies.ipynb)\n",
     "- [Simple (interactive) time series plot](Plotting/TimeSeries.ipynb)"
    ]
   },