Cleanup: cleaned up python code

Author: ltanak

src/benchmarking/python/image_editing/diff.png

@@ -21,18 +21,22 @@ def __init__(self, image_1: Path, image_2: Path):
         if self.image_2.size != self.image_1.size:
             self.image_2 = self.image_2.resize(image_1.size)
 
+    """
+    CURRENTLY WIP
+    - DOES NOT ACHIEVE DESIRED RESULT
+    """
     def diff(self, output_name: str = "diff.png"):
         copy_1 = self.image_1
         copy_2 = self.image_2
 
         diff = ImageChops.difference(copy_1, copy_2)
 
-        overlay = Image.alpha_composite(copy_1, copy_2)
+        # overlay = Image.alpha_composite(copy_1, copy_2)
 
         script_dir = Path(__file__).resolve().parent
         output_path = script_dir / output_name
 
-        overlay.save(output_path)
+        diff.save(output_path)
 
         print(f"Saved diff to: {output_path}")
 

src/benchmarking/python/scripts/image.py …nchmarking/python/image_editing/image.pysrc/benchmarking/python/scripts/image.py renamed to src/benchmarking/python/image_editing/image.py src/benchmarking/python/scripts/image.py renamed to src/benchmarking/python/image_editing/image.py

@@ -13,4 +13,4 @@ pillow = "*"
 
 [tasks]
 main = "python -m scripts.main"
-picture = "python -m scripts.image"
+image = "python -m image_editing.image"

…/benchmarking/python/scripts/overlay.png …marking/python/image_editing/overlay.pngsrc/benchmarking/python/scripts/overlay.png renamed to src/benchmarking/python/image_editing/overlay.png src/benchmarking/python/scripts/overlay.png renamed to src/benchmarking/python/image_editing/overlay.png

@@ -42,21 +42,9 @@ def iter_rows(self):
             }
 
     # 2D array of processed data frame into  [["Open", "Close", "High", "Low", "Volume"], ...]
+    # aggregates trades into candlesticks over 'cycles' instead of timestamp
+    # last value in array is the trade price
     def _process_csv(self, cycles: int = 10, cycle_range: tuple[int, int] | None = None, source: str = "expected") -> list[list[float]]:
-        """
-        agg trades into candlesticks over `cycles` rows.
-
-        Uses the last value of prices in a row as the trade price; volume is
-        the sum of quantities for that row. Empty rows (no trades) advance the
-        cycle window but do not change OHLCV unless a trade occurs in the window.
-        Returns a list of [Open, Close, High, Low, Volume].
-        
-        Args:
-            cycles: Number of rows per candlestick window.
-            cycle_range: Optional (min, max) tuple to filter rows by cycle number.
-            source: Either "expected" (exp_prices/exp_qtities) or "actual" (acc_prices/acc_qtities).
-        """
-
         candles: list[list[float]] = []
 
         open_price = None
@@ -112,11 +100,10 @@ def _process_csv(self, cycles: int = 10, cycle_range: tuple[int, int] | None = N
 
         return candles
 
-
-
     def _split_column(self, col: str):
         if col not in self.df:
             return
+        
         def _to_float_list(value):
             if pd.isna(value):
                 return []
@@ -128,8 +115,10 @@ def _to_float_list(value):
             except TypeError:
                 return [float(value)]
 
-        self.df[col] = self.df[col].apply(_to_float_list)
-
+        self.df[col] = self.df[col].apply(lambda x: _to_float_list(x))
+        # self.df[col] = self.df[col].apply(_to_float_list) is equivalent, e.g. x -> f -> f x
+        
+    # plots the candlestick chart
     def plot_candles(self, data, start="2026-01-01", freq="T", title="Stock Price", out: Path | None = None):
         df = pd.DataFrame(
             data,
@@ -155,19 +144,12 @@ def plot_candles(self, data, start="2026-01-01", freq="T", title="Stock Price",
         )
 
     def plot_all(self, out_dir: Path | None = None, cycle_range: tuple[int, int] | None = None):
-        """
-        Generate and save exchange candlestick plots for both expected and actual data.
-        
-        Args:
-            out_dir: Output directory. If None, uses default exchange graphs directory.
-            cycle_range: Optional (min, max) tuple to filter data by cycle range.
-        """
         if out_dir is None:
             out_dir = get_graph_dir("exchange")
 
         base_name = self.csv_path.stem
 
-        # Plot expected data
+        # process csv into useable form
         candles_expected = self._process_csv(cycle_range=cycle_range, source="expected")
         if candles_expected:
             self.plot_candles(
@@ -180,7 +162,7 @@ def plot_all(self, out_dir: Path | None = None, cycle_range: tuple[int, int] | N
         else:
             print("No expected trades found")
 
-        # Plot actual data
+        # plot actual data
         candles_actual = self._process_csv(cycle_range=cycle_range, source="actual")
         if candles_actual:
             self.plot_candles(

src/benchmarking/python/pixi.toml

@@ -14,7 +14,7 @@ def __init__(self, csv_path: Path):
             "dequeue": df["dequeue_latency_ns"].astype(float).values,
         }
 
-    # Outlier removal (IQR)
+    # outlier removal (IQR)
     def _remove_outliers_iqr(self, data, k=1.5):
         q1 = np.percentile(data, 25)
         q3 = np.percentile(data, 75)
@@ -96,7 +96,7 @@ def cdf(self, metric="enqueue", title="Latency CDF", out: Path | None = None, re
 
         plt.close()
 
-    # Summary
+    # plot summary of factors
     def summary(self, metric="enqueue", remove_outliers=True):
         data = self.data[metric]
         if remove_outliers:

src/benchmarking/python/plotting/ExchangePlot.py

@@ -2,7 +2,11 @@
 import pandas as pd
 import numpy as np
 import matplotlib.pyplot as plt
-from benchmarking.python.plotting.utils import get_graph_dir
+from benchmarking.python.plotting.utils import *
+
+"""
+Class to plot all the graphs related to order-preservation
+"""
 
 class OrderingPlot:
     def __init__(self, csv_path: Path):
@@ -14,6 +18,7 @@ def __init__(self, csv_path: Path):
             "actual_id": df["actual_id"].astype(float).values,
         }
 
+    # checks if sequence is identical
     def order_preserved(self) -> bool:
         return np.array_equal(self.data["expected_id"], self.data["actual_id"])
 
@@ -65,15 +70,15 @@ def plot_offset(self, title="Actual - Expected", out: Path | None = None, id_ran
         exp = self.data["expected_id"]
         act = self.data["actual_id"]
 
-        # Filter by id_range if provided
+        # id_range filter
         if id_range:
             mask = (exp >= id_range[0]) & (exp <= id_range[1])
             exp = exp[mask]
             act = act[mask]
 
         delta = act - exp
 
-        # Only plot points with non-zero offset
+        # plot points with non-zero offset
         offset_mask = delta != 0
         exp_offset = exp[offset_mask]
         delta_offset = delta[offset_mask]
@@ -97,7 +102,6 @@ def plot_displacement_heatmap(self, title="Displacement Magnitude", out: Path |
         exp = self.data["expected_id"]
         act = self.data["actual_id"]
 
-        # Filter by id_range if provided
         if id_range:
             mask = (exp >= id_range[0]) & (exp <= id_range[1])
             exp = exp[mask]
@@ -125,7 +129,6 @@ def plot_expected_vs_actual_colored(self, title="Expected vs Actual (colored by
         exp = self.data["expected_id"]
         act = self.data["actual_id"]
 
-        # Filter by id_range if provided
         if id_range:
             mask = (exp >= id_range[0]) & (exp <= id_range[1])
             exp = exp[mask]
@@ -150,15 +153,8 @@ def plot_expected_vs_actual_colored(self, title="Expected vs Actual (colored by
             plt.show()
         plt.close()
 
-    # Plot all ordering graphs
+    # plot all ordering graphs
     def plot_all(self, out_dir: Path | None = None, id_range: tuple[int, int] | None = None):
-        """
-        Generate and save all ordering plots to the specified directory.
-        
-        Args:
-            out_dir: Output directory. If None, uses default ordering graphs directory.
-            id_range: Optional tuple (min, max) to filter data by ID range.
-        """
         if out_dir is None:
             out_dir = get_graph_dir("ordering")
 

src/benchmarking/python/plotting/LatencyPlot.py

@@ -1,4 +1,5 @@
 from pathlib import Path
+
 """
 Code that contains all the utility functions for getting most recent graphs,
 csvs, and directory paths.
@@ -8,22 +9,24 @@
 def get_root() -> Path:
     return Path(__file__).resolve().parents[2]
 
+# returns the path for the specific csv directory (exchange, ordering, latencies)
 def get_csv_dir(dir: str) -> Path:
     return get_root() / "csvs" / dir
 
+# eturns the path for the specific graph directory (exchange, ordering, latencies)
 def get_graph_dir(dir: str) -> Path:
     path = get_root() / "graphs" / dir
     path.mkdir(parents=True, exist_ok=True)
     return path
 
-# get the most recent CSV from all directories (exchange, ordering, latencies).
+# get the most recent CSV from all directories (exchange, ordering, latencies)
 def get_latest_csv(must_match: bool = False) -> dict[str, Path]:
-    dirs = {"exchange": "exchange", "ordering": "ordering", "latencies": "latencies"}
+    dirs = ["exchange", "ordering", "latencies"]
     latest = {}
     timestamps = {}
 
-    for key, dir_name in dirs.items():
-        path = get_csv_dir(dir_name)
+    for key in dirs:
+        path = get_csv_dir(key)
         csvs = get_csvs_dir(path, reverse=True)
         if not csvs:
             raise FileNotFoundError(f"No CSVs found in {path}")
@@ -44,9 +47,11 @@ def get_latest_csv(must_match: bool = False) -> dict[str, Path]:
 
     return latest
 
+# returns the most recent csv in the directory specified
 def get_latest_csv_dir(dir: str) -> Path:
     return get_latest_csv()[dir]
 
+# returns all the csvs in the directory specified
 def get_csvs_dir(path: Path, reverse: bool = False):
     return sorted(path.glob("*.csv"), reverse=reverse)
 
@@ -60,13 +65,8 @@ def get_csv_all_dirs(name: str) -> dict[str, bool]:
     dirs = ["exchange", "ordering", "latencies"]
     return {dir_name: get_csv(name, dir_name) for dir_name in dirs}
 
+# returns a dictionary of directories to Paths if the file name is in any of the directories
 def get_csv_all_dirs_name(name: str) -> dict[str, Path]:
-    """
-    Find a CSV by name across all directories and return full paths.
-    
-    Returns:
-        Dictionary mapping dir names to full CSV paths where the file exists.
-    """
     dirs = ["exchange", "ordering", "latencies"]
     result = {}
     for dir_name in dirs: