add comments and function documentation notebook_statistic

notebooks/notebook_statistic_data.ipynb

@@ -209,7 +209,8 @@
     "import matplotlib.dates as mdates\n",
     "\n",
     "output_folder = \"../data/figures/output_fig_statistic\"\n",
-    "os.makedirs(output_folder, exist_ok=True)  # Create the folder if it doesn't exist\n",
+    "# Create the folder if it doesn't exist\n",
+    "os.makedirs(output_folder, exist_ok=True)\n",
     "\n",
     "\n",
     "temp = df['temp.mean_celsius']\n",
@@ -380,37 +381,61 @@
     "    \"winter\": winter\n",
     "    }\n",
     "\n",
-    "def check_outliers(seasons, temp):\n",
-    "    temp_mean = df[temp]\n",
+    "def check_outliers(df, seasons, temp_column):\n",
+    "    '''\n",
+    "    This function takes in the parameters season and temp. Seasons is a list of the different seasons, with masked dates.\n",
+    "    The temp, indicates either the mean temperature, the max measured temp or the lowest measured temperature for each day.\n",
+    "    The function then calculates both the mean and standard deviation based on the temp. Before calculating the upper and\n",
+    "    lower limit with 3 standard deviation away from the mean. Then it uses these values to find the outliers, which is \n",
+    "    either 3 standarddeviation above or below the mean. This is based on the seasons, so for each season, it find the outliers.\n",
+    "    The outliers are then given the nan value, before using interpolate to 'guess' the missing value. Then the outliers are dealt with. \n",
+    "    '''\n",
     "\n",
+    "    # Finds the temperature in the dataframe according to the inputed value\n",
+    "    temp = df[temp_column]\n",
+    "\n",
+    "    # Goes through each season, and calcute mean and st.dev and searching for outliers\n",
     "    for season_name, season_mask in seasons.items():\n",
-    "        temp_mean_season = temp_mean[season_mask]\n",
-    "        season_mean = temp_mean_season.mean()\n",
-    "        season_stdev = temp_mean_season.std()\n",
     "\n",
+    "        # Stores the temperature based on the season mask dates\n",
+    "        temp_season = temp[season_mask]\n",
+    "\n",
+    "        # Calculates the mean for the season\n",
+    "        season_mean = temp_season.mean()\n",
+    "\n",
+    "        # Calculates the standard deviation for the season\n",
+    "        season_stdev = temp_season.std()\n",
+    "\n",
+    "        # Calculates the limits for 3 standard deviation from the mean\n",
     "        upper_limit = season_mean + (season_stdev * 3)\n",
     "        lower_limit = season_mean - (season_stdev * 3)\n",
     "\n",
-    "        outliers_mask = (temp_mean_season > upper_limit) | (temp_mean_season < lower_limit)\n",
+    "        # Creates a mask for the temperatures above or below the upper/lower limit\n",
+    "        outliers_mask = (temp_season > upper_limit) | (temp_season < lower_limit)\n",
     "\n",
-    "        outlier_index = temp_mean_season.index[outliers_mask]\n",
+    "        # Gets the index for the outliers\n",
+    "        outlier_index = temp_season.index[outliers_mask].tolist()\n",
     "\n",
+    "        # Cheks if the 'month_day' column are in the dataframe, and then add it with the output for the outliers\n",
     "        if 'month_day' in df.columns:\n",
-    "            outliers = df.loc[outlier_index, ['month_day', temp]]\n",
+    "            outliers = df.loc[outlier_index, ['month_day', temp_column]]\n",
     "        else:\n",
-    "            outliers = df.loc[outlier_index, temp]\n",
+    "            outliers = df.loc[outlier_index, temp_column]\n",
     "\n",
+    "        # Prints the season name, before the outliers\n",
     "        print(season_name)\n",
     "        print(outliers)\n",
     "\n",
     "        # Replace with NaN\n",
-    "        df.loc[outlier_index, temp] = np.nan\n",
+    "        df.loc[outlier_index, temp_column] = np.nan\n",
     "\n",
-    "        df[temp] = df[temp].interpolate(method='linear', limit_direction='both')\n",
+    "        # Interpolate other missing 'NaN'-values\n",
+    "        df[temp_column] = df[temp_column].interpolate(method='linear', limit_direction='both')\n",
     "\n",
-    "check_outliers(seasons, 'temp.mean_celsius')\n",
-    "check_outliers(seasons, 'temp.record_min_celsius')\n",
-    "check_outliers(seasons, 'temp.max_celsius')"
+    "# Runs the funciton for each of the wanted temperature columns in the df\n",
+    "check_outliers(df, seasons, 'temp.mean_celsius')\n",
+    "check_outliers(df, seasons, 'temp.record_min_celsius')\n",
+    "check_outliers(df, seasons, 'temp.max_celsius')"
    ]
   },
   {
@@ -448,8 +473,9 @@
     "plt.plot(temp_record_min.index, temp_record_min, color='tab:blue', label='Min Temperature')\n",
     "\n",
     "# Customize the x-axis to show ticks and labels only at the start of each month\n",
-    "plt.gca().xaxis.set_major_locator(mdates.MonthLocator())  \n",
-    "plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%b'))  # Format ticks to show abbreviated month names (e.g., Jan, Feb)\n",
+    "plt.gca().xaxis.set_major_locator(mdates.MonthLocator())\n",
+    "# Format ticks to show abbreviated month names (e.g., Jan, Feb)\n",
+    "plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%b'))  \n",
     "\n",
     "plt.axhline(y=0, color='black', linewidth=1.5)\n",
     "\n",
@@ -544,7 +570,6 @@
     "# Normalize the data for better readability\n",
     "df_records = pd.json_normalize(data)\n",
     "\n",
-    "\n",
     "# Displays the dataframe\n",
     "display(df_records)"
    ]