add llm chapter

boa/analysis/aufgabe_02.py

@@ -8,11 +8,14 @@ import matplotlib.pyplot as plt
 import matplotlib
 matplotlib.use('TkAgg')
 
-JAVA_INPUT = './data/try_Java_Jan_2022_last_revision.boa.output.txt'
-PYTHON_INPUT = './data/try_Python_Feb_2022_last_revision.boa.output.txt'
+# Input Pfade
+JAVA_INPUT = './boa-output/try_Java_Jan_2022_last_revision.boa.output.txt'
+PYTHON_INPUT = './boa-output/try_Python_Feb_2022_last_revision.boa.output.txt'
 
-SHOW_PLOTS = False
-SEED = 3
+# Konfiguration von read_sample
+SEED = 1
+SHOW_PLOTS = True
+SAMPLES = 1000
 
 def read_sample(file_path, title='', sample_size=0, show=True):
     # filename kann auch ein URL sein: "https://..../example.csv"
@@ -50,12 +53,11 @@ def cohen_d(x, y):
     # Stichprobengrößen der Gruppen
     size_x, size_y = len(x), len(y)
 
-    # Gepoolte Varianz und Standardabweichung
-    pool_var = ((size_x - 1) * var_x + (size_y - 1) * var_y) / (size_x + size_y - 2)
-    pool_var = math.sqrt(pool_var)
+    # Gepoolte Standardabweichung
+    pool_std = math.sqrt(((size_x - 1) * var_x + (size_y - 1) * var_y) / (size_x + size_y - 2))
 
     # Cohen's d
-    d_val = mean_diff / pool_var
+    d_val = mean_diff / pool_std
     return d_val
 
 def find_barrier(x_sample_path, y_sample_path, alpha=0.1, lower_limit=2, upper_limit=10000):
@@ -95,12 +97,12 @@ def main():
     print('Statistische Berechnungen zu Häufigkeiten (Übung 5)')
 
     print('\nEinlesen der ersten Stichprobe (Python)')
-    python_sample = read_sample(PYTHON_INPUT, sample_size=1000, show=SHOW_PLOTS, title="Python")
+    python_sample = read_sample(PYTHON_INPUT, sample_size=SAMPLES, show=SHOW_PLOTS, title="Python")
     print('Mean:' + str(python_sample['Ratio'].mean()))
     print('Variance:' + str(python_sample['Ratio'].var()))
 
     print('\nEinlesen der zweiten Stichprobe (Java)')
-    java_sample = read_sample(JAVA_INPUT, sample_size=1000, show=SHOW_PLOTS, title="Java")
+    java_sample = read_sample(JAVA_INPUT, sample_size=SAMPLES, show=SHOW_PLOTS, title="Java")
     print('Mean:' + str(java_sample['Ratio'].mean()))
     print('Variance:' + str(java_sample['Ratio'].var()))
 

boa/analysis/test.py

@@ -1,9 +0,0 @@
-import random
-
-random.seed(1)
-print(random.randint(0, 10))
-print(random.randint(0, 10))
-print(random.randint(0, 10))
-print(random.randint(0, 10))
-print(random.randint(0, 10))
-print(random.randint(0, 10))

boa/mining/aufgabe_01_a.boa

@@ -1,5 +1,5 @@
-# Java-Job: 111005
-# Python-Job: 111006
+# Java-Job: 111839
+# Python-Job: 111840
 
 # Includes all revisions
 p: Project = input;

boa/mining/aufgabe_01_b.boa

@@ -1,5 +1,5 @@
-# Java-Job: 111694
-# Python-Job: 111439
+# Java-Job: Throws empty error on the "2022 Jan/Java" dataset.
+# Python-Job: 111838
 
 # Includes all revisions
 p: Project = input;
@@ -12,7 +12,7 @@ cur_date: time;
 
 statement_counter := visitor {
     before node: Statement -> {
-        if (def(node.kind) and (node.kind == StatementKind.TRY or node.kind == StatementKind.WITH))
+        if (node.kind == StatementKind.TRY or node.kind == StatementKind.WITH)
             relative_list[p.name][yearof(cur_date)] << 1;
         else
             relative_list[p.name][yearof(cur_date)] << 0;
@@ -21,8 +21,8 @@ statement_counter := visitor {
 
 visit(p, visitor {
     before node: CodeRepository -> {
-        for (minus_year: int=22; minus_year >= 0; minus_year--) {
-            cur_date = addyear(T"Dec 31, 2022, 10:00:00 AM", -minus_year);
+        for (minus_year: int=19; minus_year >= 0; minus_year--) {
+            cur_date = addyear(T"Dec 31, 2019, 10:00:00 AM", -minus_year);
             snapshot := getsnapshot(node, cur_date);
 
             foreach (i: int; def(snapshot[i]))