label new

create_triplets.py

@@ -10,15 +10,15 @@ def create_triplets(corpus_path="data/esg_corpus.csv", output_path="data/esg_tri
     
     try:
         df = pd.read_csv(corpus_path)
-        if 'text' not in df.columns:
-            print(f"Error: Corpus file at {corpus_path} must have a 'text' column.")
+        if 'markdown' not in df.columns:
+            print(f"Error: Corpus file at {corpus_path} must have a 'markdown' column.")
             return
     except FileNotFoundError:
         print(f"Error: Processed corpus file not found at {corpus_path}.")
         print("Please run the preprocessing step in the main application first.")
         return
 
-    sentences = df['text'].dropna().tolist()
+    sentences = df['markdown'].dropna().tolist()
     if len(sentences) < 3:
         print("Error: Not enough sentences in the corpus to generate triplets.")
         return