PerSent 0.2.2__tar.gz → 0.2.4__tar.gz

{persent-0.2.2 → persent-0.2.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PerSent
-Version: 0.2.2
+Version: 0.2.4
 Summary: Persian Sentiment Analysis Toolkit
 Home-page: https://github.com/RezaGooner/PerSent
 Author: RezaGooner

{persent-0.2.2 → persent-0.2.4}/PerSent/CommentAnalyzer.py

@@ -112,40 +112,88 @@ class CommentAnalyzer:
         self.classifier = joblib.load(os.path.join(self.model_dir, 'classifier.joblib'))
         self.vectorizer = Word2Vec.load(os.path.join(self.model_dir, 'word2vec.model'))
         
-    def csvPredict(self, input_csv, output_path, text_column='body'):
+    def csvPredict(self, input_csv, output_path, summary_path=None, text_column=0):
         """
         Analyze sentiment for comments in a CSV file and save results
         
         Parameters:
             input_csv (str): Path to input CSV file
             output_path (str): Path to save output CSV file
-            text_column (str, optional): Name of column containing comments. Defaults to 'body'.
+            text_column (str/int, optional): Name or index (0-based) of column containing comments. 
+                                          Defaults to 0 (first column).
+            summary_path (str, optional): Path to save prediction summary report.
+                                       If None, no summary will be saved.
         """
         try:
             # Read input CSV
             df = pd.read_csv(input_csv)
             
-            # Check if specified column exists
-            if text_column not in df.columns:
-                # If only one column exists, use it regardless of name
-                if len(df.columns) == 1:
-                    text_column = df.columns[0]
-                else:
+            # Determine the correct column
+            if isinstance(text_column, int):
+                # Handle negative indices
+                if text_column < 0:
+                    text_column = len(df.columns) + text_column
+                    
+                if text_column >= len(df.columns) or text_column < 0:
+                    raise ValueError(f"Column index {text_column} is out of range")
+                    
+                column_name = df.columns[text_column]
+            else:
+                if text_column not in df.columns:
                     raise ValueError(f"Column '{text_column}' not found in CSV file")
+                column_name = text_column
             
             # Analyze each comment
             tqdm.pandas(desc="Analyzing comments")
-            df['sentiment'] = df[text_column].progress_apply(self.predict)
+            df['sentiment'] = df[column_name].progress_apply(self.predict)
             
             # Save results
             df.to_csv(output_path, index=False, encoding='utf-8-sig')
             print(f"Results saved to {output_path}")
             
+            # Generate and save summary if requested
+            if summary_path:
+                summary = self._generate_summary(df)
+                summary.to_csv(summary_path, index=False, encoding='utf-8-sig')
+                print(f"Summary report saved to {summary_path}")
+            
             return df
             
         except Exception as e:
             print(f"Error: {str(e)}")
             return None
 
+    def _generate_summary(self, df):
+        """Generate prediction summary statistics"""
+        # Count each sentiment
+        counts = df['sentiment'].value_counts().to_dict()
+        
+        # Create summary dataframe
+        summary = pd.DataFrame({
+            'Category': [
+                'Recommended',
+                'Not Recommended', 
+                'No Idea',
+                'Total',
+                'Model Accuracy'
+            ],
+            'Count': [
+                counts.get('recommended', 0),
+                counts.get('not_recommended', 0),
+                counts.get('no_idea', 0),
+                len(df),
+                'N/A'  # Accuracy needs to be calculated during training
+            ],
+            'Percentage': [
+                f"{100 * counts.get('recommended', 0) / len(df):.2f}%",
+                f"{100 * counts.get('not_recommended', 0) / len(df):.2f}%",
+                f"{100 * counts.get('no_idea', 0) / len(df):.2f}%",
+                '100%',
+                'N/A'
+            ]
+        })
+        
+        return summary
+
 
 # Github : RezaGooner

{persent-0.2.2 → persent-0.2.4}/PerSent/__init__.py

@@ -1,4 +1,4 @@
 from .CommentAnalyzer import CommentAnalyzer
 
-__version__ = "0.2.2"
+__version__ = "0.2.4"
 __all__ = ['CommentAnalyzer']

{persent-0.2.2 → persent-0.2.4}/PerSent.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PerSent
-Version: 0.2.2
+Version: 0.2.4
 Summary: Persian Sentiment Analysis Toolkit
 Home-page: https://github.com/RezaGooner/PerSent
 Author: RezaGooner

{persent-0.2.2 → persent-0.2.4}/setup.py

@@ -5,7 +5,7 @@ with open("README.md", "r", encoding="utf-8") as fh:
 
 setup(
     name="PerSent",
-    version="0.2.2",
+    version="0.2.4",
     author="RezaGooner",
     author_email="RezaAsadiProgrammer@Gmail.com",
     description="Persian Sentiment Analysis Toolkit",