data into code and summary printing

testanalysis.py

@@ -0,0 +1,87 @@
+import os
+from collections import Counter
+
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+
+def linesplit(line):
+    """
+    Splits line of file into useful components.
+
+    Returns (dm, pulse width, originating filename).
+    """
+
+    dm, line = line.split("pc/cc")
+    dm = int(dm)
+    pw, fname = line.split(" s ")
+    pw = float(pw)
+    fname = fname.strip().removesuffix("_injected")
+    return (dm, pw, fname)
+
+def updateDic(lines, dic):
+    """
+    Processes a list of lines and adds them to given dictionary in-place.
+    """
+
+    for line in lines:
+        dm, pw, fname = linesplit(line)
+        dic["file"].append(fname)
+        dic["dm"].append(dm)
+        dic["pulseWidth"].append(pw)
+
+injectedDic = {
+    "file" : [],
+    "dm" : [],
+    "pulseWidth" : []
+}
+
+detectedDic = {
+    "file" : [],
+    "dm" : [],
+    "pulseWidth" : []
+}
+
+#load injected data into dataframe
+with open(os.path.join(".","out","2025-07-31T17-25-54.txt"), "r") as file:
+    lines = file.readlines()
+    updateDic(lines, injectedDic)
+
+injections = pd.DataFrame(data=injectedDic) #this is our main object for injection data
+
+#load detection data into dataframe
+with open(os.path.join(".","out","plotOut.txt"), "r") as file:
+    lines = file.readlines()
+    updateDic(lines, detectedDic)
+
+detections = pd.DataFrame(data=detectedDic) #this is our main object for detection data
+
+#define summary printing for multiple steps
+def summary(stage):
+    print(f"Summary Stage {stage}")
+    print(f"Number of files injected:        {len(Counter(injections['file']))}")
+    print(f"Number of files with detections: {len(Counter(detections['file']))}")
+    print("=========================")
+    print(f"Number of pulses injected: {len(injections['dm'])}")
+    print(f"Number of pulses detected: {len(detections['dm'])}")
+    print("=========================")
+    print(f"File ratio:      {round(len(Counter(detections['file']))/len(Counter(injections['file'])), 3)}")
+    print(f"Detection ratio: {round(len(detections['dm'])/len(injections['dm']), 3)}")
+    print("=========================")
+
+#print initial summary
+summary(1)
+
+#many files contained pulsar bursts, so we filter those out via DM
+minDM = (10**2.5) * 0.95 #as per signal generation, plus a bit of wiggle room
+print(f"Filtering out pulsars (DM below {int(minDM)}...)")
+detections = detections[detections['dm'] > minDM]
+summary(2)
+
+#Let's do detection matching! Yaaaay!
+#What detections line up to which injections? This will determine which ones got missed entirely.
+#Define some kind of epsilon for DM and pulse width; if detection is within epsilon in both DM and PW we can match it.
+dmEps = 0.05
+pwEps = 0.2
+#and define an auxiliary array of bools for injections containing the "is matched?" information.
+isMatched = [False] * len(injections['dm'])