update homework

RequirementsTraceabilityAssignment/sorted_frs.txt

@@ -0,0 +1,60 @@
+FR2: 0.7272982597351074
+FR59: 0.6924091577529907
+FR41: 0.658149242401123
+FR3: 0.6336919665336609
+FR32: 0.6048915386199951
+FR9: 0.5977640151977539
+FR13: 0.5880354046821594
+FR30: 0.5614917278289795
+FR6: 0.5527222156524658
+FR56: 0.501095712184906
+FR35: 0.4926453232765198
+FR51: 0.4922177791595459
+FR43: 0.4850791394710541
+FR27: 0.4834768772125244
+FR50: 0.4728085994720459
+FR16: 0.4689098000526428
+FR23: 0.46860480308532715
+FR10: 0.4604903757572174
+FR42: 0.4555082321166992
+FR47: 0.4390988349914551
+FR40: 0.4389651417732239
+FR49: 0.43022799491882324
+FR54: 0.4262740910053253
+FR21: 0.41903120279312134
+FR15: 0.40128734707832336
+FR28: 0.39869096875190735
+FR1: 0.372334361076355
+FR7: 0.36662811040878296
+FR19: 0.3618888854980469
+FR5: 0.36042946577072144
+FR14: 0.3450104296207428
+FR44: 0.3450104296207428
+FR34: 0.3366490602493286
+FR33: 0.33334457874298096
+FR58: 0.3215200901031494
+FR60: 0.28449445962905884
+FR24: 0.282522588968277
+FR29: 0.27975350618362427
+FR31: 0.27687233686447144
+FR36: 0.2559022009372711
+FR20: 0.25007960200309753
+FR18: 0.2497757077217102
+FR57: 0.24898050725460052
+FR46: 0.23702991008758545
+FR45: 0.23629112541675568
+FR4: 0.2229391634464264
+FR53: 0.21678310632705688
+FR52: 0.2120310366153717
+FR22: 0.20914143323898315
+FR39: 0.20602846145629883
+FR17: 0.20446419715881348
+FR8: 0.20398864150047302
+FR48: 0.2015649676322937
+FR26: 0.18092623353004456
+FR12: 0.17228464782238007
+FR25: 0.16649314761161804
+FR11: 0.14829087257385254
+FR37: 0.13823707401752472
+FR55: 0.13276946544647217
+FR38: 0.09154436737298965

RequirementsTraceabilityAssignment/traceability_fergujp.py

@@ -278,6 +278,8 @@ def run_option1(filepath, answerpath):
 
     sim = tfidf_similarity(frs, nfrs)
 
+    sort_similarities(frs, sim)
+
     THRESHOLDS_TFIDF = [0.1, 0.1, 0.1]
 
     preds = predict_labels(sim, thresholds=THRESHOLDS_TFIDF)
@@ -290,6 +292,8 @@ def run_option2(filepath, answerpath):
 
     sim = embedding_similarity(frs, nfrs)
 
+    sort_similarities(frs, sim)
+
     THRESHOLDS_EMBEDDING = [0.3, 0.3, 0.3]
 
     preds = predict_labels(sim, thresholds=THRESHOLDS_EMBEDDING)
@@ -310,7 +314,64 @@ def run_option3(filepath, answerpath):
     evaluate(preds, frs, answers)
 
 # ============================================================
-# 9. RUN PIPELINE
+# 10. OPTIONAL MERGE SORTING OF FRs BASED ON SIMILARITY SCORES
+# ============================================================
+def sort_similarities(frs, similarities):
+    """
+    Sorts FRs based on max similarity to any NFR
+    TODO - compared to each NFR separately
+    """
+
+    fr_sim_pairs = [(fr_id, max(similarities[i])) for i, (fr_id, _) in enumerate(frs)]
+
+    # Use manual merge sort instead of sorted()
+    sorted_pairs = merge_sort(fr_sim_pairs)
+
+    # TODO - output as text with the actual FR texts
+    with open("sorted_frs.txt", "w") as f:
+        for fr_id, sim in sorted_pairs:
+            f.write(f"{fr_id}: {sim}\n")
+
+    return sorted_pairs
+
+def merge_sort(fr_sim_pairs):
+    if len(fr_sim_pairs) <= 1:
+        return fr_sim_pairs
+
+    mid = len(fr_sim_pairs) // 2
+    left = merge_sort(fr_sim_pairs[:mid])
+    right = merge_sort(fr_sim_pairs[mid:])
+
+    return merge(left, right)
+
+
+def merge(left, right):
+    result = []
+    i = 0
+    j = 0
+
+    # Descending order by similarity (index 1)
+    while i < len(left) and j < len(right):
+        if left[i][1] >= right[j][1]:
+            result.append(left[i])
+            i += 1
+        else:
+            result.append(right[j])
+            j += 1
+
+    while i < len(left):
+        result.append(left[i])
+        i += 1
+
+    while j < len(right):
+        result.append(right[j])
+        j += 1
+
+    return result
+
+
+# ============================================================
+# 11. RUN PIPELINE
 # ============================================================
 if __name__ == "__main__":
     run_option1(DATAPATH, ANSWERPATH)