[M] Mode experiment scripts

experiments/action_sizes.py

@@ -0,0 +1,69 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from pathlib import Path
+
+import numpy as np
+import matplotlib.pyplot as plt
+
+@dataclass
+class Statistics:
+    mean: float
+    median: float
+    lower_quartile: float
+    upper_quartile: float
+    iqr: float
+    minimum: float
+    maximum: float
+    count: int
+    total: float
+    stddev: float
+
+    def get_metric_6(self) -> tuple[float, float, float, float, float, float]:
+        return self.mean, self.median, self.minimum, self.maximum, self.lower_quartile, self.upper_quartile
+
+    def print(self, dec: int = 2):
+        print(f'> Mean: {round(self.mean, dec)}, Median: {round(self.median, dec)}')
+        print(f'> Min: {round(self.minimum, dec)}, Max: {round(self.maximum, dec)}')
+        print(f'> Q1: {round(self.lower_quartile, dec)}, Q3: {round(self.upper_quartile, dec)}')
+        print(f'> StdDev: {round(self.stddev, dec)}, IQR: {round(self.iqr, dec)}')
+        print(f'> N: {self.count}')
+
+
+def _calc_col_stats_helper(col: np.ndarray) -> tuple[float, float, float, float, float, float, float, int, float, float]:
+    q1 = np.quantile(col, 0.25)
+    q3 = np.quantile(col, 0.75)
+    return (
+        float(np.mean(col)),
+        float(np.median(col)),
+        float(q1),
+        float(q3),
+        float(q3 - q1),
+        float(np.min(col)),
+        float(np.max(col)),
+        len(col),
+        float(np.sum(col)),
+        float(np.std(col))
+    )
+
+
+def calc_col_stats(col: np.ndarray | list) -> Statistics:
+    """
+    Compute statistics for a data column
+    :param col: Input column (tested on 1D array)
+    :return: Statistics
+    """
+    if isinstance(col, list):
+        col = np.array(col)
+    return Statistics(*_calc_col_stats_helper(col))
+
+
+if __name__ == '__main__':
+    txt = Path('action-sizes.log').read_text('utf-8').split('\n')
+    nums = [int(line) for line in txt if line.isnumeric()]
+    # print(nums)
+
+    calc_col_stats(nums).print()
+    plt.hist(nums, bins=50)
+    plt.show()
+