PerSent 0.1.6__tar.gz → 0.1.8__tar.gz

persent-0.1.8/PKG-INFO

@@ -0,0 +1,28 @@
+Metadata-Version: 2.4
+Name: PerSent
+Version: 0.1.8
+Summary: Persian Sentiment Analysis Toolkit
+Home-page: https://github.com/RezaGooner/PerSent
+Author: RezaGooner
+Author-email: RezaAsadiProgrammer@Gmail.com
+Keywords: persian sentiment analysis nlp
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+Requires-Dist: hazm>=0.7.0
+Requires-Dist: gensim>=4.0.0
+Requires-Dist: scikit-learn>=1.0.0
+Requires-Dist: pandas>=1.3.0
+Requires-Dist: tqdm>=4.62.0
+Requires-Dist: joblib>=1.1.0
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: keywords
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary

persent-0.1.8/PerSent.egg-info/PKG-INFO

@@ -0,0 +1,28 @@
+Metadata-Version: 2.4
+Name: PerSent
+Version: 0.1.8
+Summary: Persian Sentiment Analysis Toolkit
+Home-page: https://github.com/RezaGooner/PerSent
+Author: RezaGooner
+Author-email: RezaAsadiProgrammer@Gmail.com
+Keywords: persian sentiment analysis nlp
+Classifier: Programming Language :: Python :: 3
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+Requires-Dist: hazm>=0.7.0
+Requires-Dist: gensim>=4.0.0
+Requires-Dist: scikit-learn>=1.0.0
+Requires-Dist: pandas>=1.3.0
+Requires-Dist: tqdm>=4.62.0
+Requires-Dist: joblib>=1.1.0
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: keywords
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary

{persent-0.1.6 → persent-0.1.8}/PerSent.egg-info/SOURCES.txt

@@ -1,11 +1,7 @@
-LICENSE.txt
 README.md
 setup.py
-PerSent/__init__.py
-PerSent/analyzer.py
 PerSent.egg-info/PKG-INFO
 PerSent.egg-info/SOURCES.txt
 PerSent.egg-info/dependency_links.txt
 PerSent.egg-info/requires.txt
-PerSent.egg-info/top_level.txt
-tests/test_basic.py
+PerSent.egg-info/top_level.txt

persent-0.1.8/PerSent.egg-info/top_level.txt

@@ -0,0 +1 @@
+

{persent-0.1.6 → persent-0.1.8}/setup.py

@@ -5,7 +5,7 @@ with open("README.md", "r", encoding="utf-8") as fh:
 
 setup(
     name="PerSent",
-    version="0.1.6",
+    version="0.1.8",
     author="RezaGooner",
     author_email="RezaAsadiProgrammer@Gmail.com",
     description="Persian Sentiment Analysis Toolkit",
@@ -31,4 +31,4 @@ setup(
     ],
     python_requires='>=3.8',
     keywords='persian sentiment analysis nlp',
-)
+)

persent-0.1.6/LICENSE.txt

@@ -1,21 +0,0 @@
-MIT License
-
-Copyright (c) 2025 𝐑𝐞𝐳𝐚 𝐀𝐬𝐚𝐝𝐢
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.

persent-0.1.6/PKG-INFO

@@ -1,162 +0,0 @@
-Metadata-Version: 2.4
-Name: PerSent
-Version: 0.1.6
-Summary: Persian Sentiment Analysis Toolkit
-Home-page: https://github.com/RezaGooner/PerSent
-Author: RezaGooner
-Author-email: RezaAsadiProgrammer@Gmail.com
-Keywords: persian sentiment analysis nlp
-Classifier: Programming Language :: Python :: 3
-Classifier: License :: OSI Approved :: MIT License
-Classifier: Operating System :: OS Independent
-Requires-Python: >=3.8
-Description-Content-Type: text/markdown
-License-File: LICENSE.txt
-Requires-Dist: hazm>=0.7.0
-Requires-Dist: gensim>=4.0.0
-Requires-Dist: scikit-learn>=1.0.0
-Requires-Dist: pandas>=1.3.0
-Requires-Dist: tqdm>=4.62.0
-Requires-Dist: joblib>=1.1.0
-Dynamic: author
-Dynamic: author-email
-Dynamic: classifier
-Dynamic: description
-Dynamic: description-content-type
-Dynamic: home-page
-Dynamic: keywords
-Dynamic: license-file
-Dynamic: requires-dist
-Dynamic: requires-python
-Dynamic: summary
-
-# Persian Sentiment Analyzer
-
-[![Python Version](https://img.shields.io/badge/python-3.6%2B-blue)](https://www.python.org/)
-[![License](https://img.shields.io/badge/license-MIT-green)](LICENSE)
-
-A Python library for sentiment analysis of Persian (Farsi) text, capable of classifying opinions as "recommended", "not_recommended", or "no_idea".
-
-## Features
-
-- **Text Preprocessing**: Normalization, tokenization, stemming, and stopword removal for Persian text
-- **Word Embeddings**: Built-in Word2Vec implementation for Persian language
-- **Sentiment Classification**: Logistic Regression classifier trained on Persian sentiment data
-- **Model Persistence**: Save and load trained models for future use
-- **Batch Processing**: Analyze sentiment for multiple texts at once
-
-## Installation
-
-```bash
-pip install persian-sentiment-analyzer
-```
-
-## Dependencies
-- Python 3.6+
-
-- hazm
-
-- gensim
-
-- scikit-learn
-
-- numpy
-
-- pandas
-
-## Usage
-
-### Basic Usage
-
-``` bash
-from persian_sentiment_analyzer import SentimentAnalyzer
-
-# Initialize with a pre-trained model
-analyzer = SentimentAnalyzer(model_path="path/to/pretrained_model")
-
-# Predict sentiment
-result = analyzer.predict("این محصول بسیار عالی است")
-print(result)  # Output: 'recommended'
-```
-
-### Training Your Own Model
-
-``` bash
-from persian_sentiment_analyzer import SentimentAnalyzer
-import pandas as pd
-
-# Load your dataset
-data = pd.read_csv("persian_reviews.csv")
-texts = data['text'].tolist()
-labels = data['label'].values  # 0: not_recommended, 1: recommended, 2: no_idea
-
-# Initialize analyzer
-analyzer = SentimentAnalyzer()
-
-# Preprocess and tokenize texts
-tokenized_texts = [analyzer.preprocessor.preprocess_text(text) for text in texts]
-
-# Train Word2Vec model
-analyzer.train_word2vec(tokenized_texts, vector_size=100)
-
-# Prepare feature vectors
-X = np.array([analyzer.sentence_vector(tokens) for tokens in tokenized_texts])
-
-# Train classifier
-analyzer.train_classifier(X, labels)
-
-# Save the trained model
-analyzer.save_model("my_persian_model")
-```
-
-### Batch Processing
-
-```bash
-from persian_sentiment_analyzer import predict_sentiments_for_file
-
-# Process a CSV file containing Persian comments
-results_summary = predict_sentiments_for_file(
-    analyzer,
-    input_file="comments.csv",
-    output_file="results.csv",
-    summary_file="summary.csv"
-)
-
-print(results_summary)
-```
-
-## Model Architecture
-1- Text Preprocessing:
-
-- Normalization (Hazm)
-
-- Tokenization
-
-- Stemming
-
-- Stopword removal
-
-2- Feature Extraction:
-
-- Word2Vec embeddings (100 dimensions)
-
-- Sentence vectors (average of word vectors)
-
-3- Classification:
-
-- Logistic Regression with L2 regularization
-
-## Performance
-The pre-trained model achieves the following performance on our test set:
-
-Metric	Value
-Accuracy	85.2%
-Precision	84.7%
-Recall	85.0%
-F1-score	84.8%
-
-License
-This project is licensed under the MIT [License](https://github.com/RezaGooner/Sentiment-Survey-Analyzer/blob/main/LICENSE) - see the LICENSE file for details
-
-
-> Github : RezaGooner

persent-0.1.6/PerSent/__init__.py

@@ -1,4 +0,0 @@
-from .analyzer import PersianSentimentAnalyzer
-
-__version__ = "0.1.6"
-__all__ = ['PersianSentimentAnalyzer']

persent-0.1.6/PerSent/analyzer.py

@@ -1,251 +0,0 @@
-import pandas as pd
-from hazm import Normalizer, word_tokenize, Stemmer, stopwords_list
-import re
-from tqdm import tqdm
-from gensim.models import Word2Vec
-import numpy as np
-from sklearn.linear_model import LogisticRegression
-from sklearn.model_selection import train_test_split
-import os
-import joblib
-from transformers import pipeline
-import torch
-from torch.nn.functional import softmax
-
-class AdvancedPersianSentimentAnalyzer:
-    def __init__(self, model_dir='persian_sentiment/model'):
-        self.normalizer = Normalizer()
-        self.stemmer = Stemmer()
-        self.stopwords = set(stopwords_list())
-        self.model_dir = model_dir
-        self.vectorizer = None
-        self.classifier = None
-        
-        # تنظیمات پیشرفته برای مدل احساسات
-        self.emotion_analyzer = pipeline(
-            "text-classification",
-            model="HooshvareLab/bert-fa-base-uncased-emotion",
-            tokenizer="HooshvareLab/bert-fa-base-uncased-emotion",
-            device=0 if torch.cuda.is_available() else -1,
-            return_all_scores=True
-        )
-        
-        # پارامترهای بهینه شده
-        self.w2v_params = {
-            'vector_size': 300,
-            'window': 10,
-            'min_count': 2,
-            'workers': 4,
-            'sg': 1  # استفاده از skip-gram
-        }
-        
-        # نگاشت برچسب‌های احساسات
-        self.emotion_labels = {
-            'ترس': 'Fear',
-            'تعجب': 'Surprise',
-            'خشم': 'Anger',
-            'تحقیر': 'Contempt',
-            'شادی': 'Joy',
-            'غم': 'Sadness',
-            'انزجار': 'Disgust'
-        }
-        
-        os.makedirs(self.model_dir, exist_ok=True)
-
-    def _preprocess_text(self, text, for_emotion=False):
-        """پیش‌پردازش متن با قابلیت تنظیم برای تحلیل احساسات"""
-        text = self.normalizer.normalize(str(text))
-        
-        if for_emotion:
-            # حفظ علائم نگارشی برای تحلیل احساسات
-            text = re.sub(r'[۰-۹\d]+', ' ', text)
-        else:
-            text = re.sub(r'[!()-\[\]{};:\'",؟<>./?@#$%^&*_~۰-۹\d]+', ' ', text)
-            
-        text = re.sub(r'\s+', ' ', text).strip()
-        
-        if not for_emotion:
-            tokens = word_tokenize(text)
-            return [
-                self.stemmer.stem(token)
-                for token in tokens
-                if token not in self.stopwords and len(token) > 1
-            ]
-        return text
-
-    def _sentence_vector(self, sentence, model):
-        """تبدیل جمله به بردار با میانگین وزنی"""
-        vectors = []
-        weights = []
-        
-        for word in sentence:
-            try:
-                vectors.append(model.wv[word])
-                # وزن‌دهی بر اساس فرکانس کلمه
-                weights.append(model.wv.get_vecattr(word, "count"))
-            except KeyError:
-                continue
-                
-        if vectors:
-            weights = np.array(weights) / sum(weights)
-            return np.average(vectors, axis=0, weights=weights)
-        return np.zeros(self.w2v_params['vector_size'])
-
-    def _emotion_analysis(self, text):
-        """تحلیل احساسات پیشرفته با مدل BERT"""
-        try:
-            # پیش‌پردازش متن برای تحلیل احساسات
-            processed_text = self._preprocess_text(text, for_emotion=True)
-            
-            # تحلیل احساسات
-            results = self.emotion_analyzer(processed_text)[0]
-            
-            # پردازش نتایج
-            emotions = {}
-            for item in results:
-                fa_label = item['label']
-                if fa_label in self.emotion_labels:
-                    en_label = self.emotion_labels[fa_label]
-                    score = round(float(item['score']), 4)
-                    emotions[en_label] = score
-            
-            # اگر همه مقادیر صفر بودند، قوی‌ترین احساس را برگردان
-            if all(v == 0 for v in emotions.values()) and results:
-                top_emotion = max(results, key=lambda x: x['score'])
-                fa_label = top_emotion['label']
-                if fa_label in self.emotion_labels:
-                    en_label = self.emotion_labels[fa_label]
-                    emotions[en_label] = round(float(top_emotion['score']), 4)
-            
-            return emotions
-            
-        except Exception as e:
-            print(f"خطا در تحلیل احساسات: {str(e)}")
-            return {label: 0.0 for label in self.emotion_labels.values()}
-
-    def train(self, train_csv, test_size=0.2):
-        """آموزش پیشرفته مدل تحلیل توصیه"""
-        df = pd.read_csv(train_csv)
-        
-        # تعادل کلاس‌ها
-        class_counts = df['recommendation_status'].value_counts()
-        max_samples = class_counts.max()
-        balanced_df = pd.concat([
-            df[df['recommendation_status'] == cls].sample(max_samples, replace=True)
-            for cls in class_counts.index
-        ])
-        
-        balanced_df['tokens'] = balanced_df['body'].progress_apply(
-            lambda x: self._preprocess_text(x))
-        
-        # آموزش مدل Word2Vec با پارامترهای بهینه
-        self.vectorizer = Word2Vec(
-            sentences=balanced_df['tokens'],
-            **self.w2v_params
-        )
-        
-        # استخراج ویژگی‌های پیشرفته
-        X = np.array([self._sentence_vector(s, self.vectorizer) 
-                     for s in tqdm(balanced_df['tokens'], desc="Vectorizing")])
-        y = balanced_df['recommendation_status'].map({
-            "no_idea": 2,
-            "recommended": 1,
-            "not_recommended": 0
-        }).values
-        
-        # تقسیم داده‌ها با stratify
-        X_train, X_test, y_train, y_test = train_test_split(
-            X, y, test_size=test_size, stratify=y)
-        
-        # مدل طبقه‌بندی با تنظیمات بهینه
-        self.classifier = LogisticRegression(
-            max_iter=2000,
-            class_weight='balanced',
-            C=0.5,
-            solver='saga'
-        )
-        self.classifier.fit(X_train, y_train)
-        
-        # ذخیره مدل
-        self.save_model()
-        
-        # ارزیابی دقیق‌تر
-        train_acc = self.classifier.score(X_train, y_train)
-        test_acc = self.classifier.score(X_test, y_test)
-        
-        return {
-            'train_accuracy': round(train_acc, 4),
-            'test_accuracy': round(test_acc, 4),
-            'model_params': self.w2v_params
-        }
-
-    def predict(self, text, mode=0, threshold=0.1):
-        """
-        پیش‌بینی پیشرفته با قابلیت تنظیم آستانه
-        
-        پارامترها:
-            text (str): متن ورودی
-            mode (int): حالت تحلیل (0=توصیه، 1=احساسات)
-            threshold (float): آستانه اطمینان برای تحلیل احساسات
-            
-        بازگشتی:
-            dict یا str: نتایج تحلیل
-        """
-        if mode == 1:
-            emotions = self._emotion_analysis(text)
-            # فیلتر کردن احساسات با احتمال پایین
-            return {
-                k: v for k, v in emotions.items() 
-                if v >= threshold
-            }
-            
-        if not self.classifier or not self.vectorizer:
-            raise Exception("Model not trained! Call train() first.")
-            
-        tokens = self._preprocess_text(text)
-        vector = self._sentence_vector(tokens, self.vectorizer)
-        proba = self.classifier.predict_proba([vector])[0]
-        
-        # تصمیم‌گیری با در نظر گرفتن احتمال
-        if max(proba) < 0.6:  # آستانه اطمینان
-            return "no_idea"
-            
-        prediction = self.classifier.predict([vector])[0]
-        return {
-            0: "not_recommended",
-            1: "recommended",
-            2: "no_idea"
-        }[prediction]
-
-    def save_model(self):
-        """ذخیره مدل با نسخه‌بندی"""
-        version = 1
-        while os.path.exists(os.path.join(self.model_dir, f'classifier_v{version}.joblib')):
-            version += 1
-            
-        joblib.dump(
-            self.classifier, 
-            os.path.join(self.model_dir, f'classifier_v{version}.joblib')
-        )
-        self.vectorizer.save(
-            os.path.join(self.model_dir, f'word2vec_v{version}.model')
-        )
-
-    def load_model(self, version=None):
-        """بارگذاری مدل با قابلیت انتخاب نسخه"""
-        if version is None:
-            versions = [
-                int(f.split('_v')[-1].split('.')[0])
-                for f in os.listdir(self.model_dir) 
-                if f.startswith('classifier_v')
-            ]
-            version = max(versions) if versions else None
-            
-        if version:
-            self.classifier = joblib.load(
-                os.path.join(self.model_dir, f'classifier_v{version}.joblib')
-            )
-            self.vectorizer = Word2Vec.load(
-                os.path.join(self.model_dir, f'word2vec_v{version}.model')
-            )
-

persent-0.1.6/PerSent.egg-info/PKG-INFO

@@ -1,162 +0,0 @@
-Metadata-Version: 2.4
-Name: PerSent
-Version: 0.1.6
-Summary: Persian Sentiment Analysis Toolkit
-Home-page: https://github.com/RezaGooner/PerSent
-Author: RezaGooner
-Author-email: RezaAsadiProgrammer@Gmail.com
-Keywords: persian sentiment analysis nlp
-Classifier: Programming Language :: Python :: 3
-Classifier: License :: OSI Approved :: MIT License
-Classifier: Operating System :: OS Independent
-Requires-Python: >=3.8
-Description-Content-Type: text/markdown
-License-File: LICENSE.txt
-Requires-Dist: hazm>=0.7.0
-Requires-Dist: gensim>=4.0.0
-Requires-Dist: scikit-learn>=1.0.0
-Requires-Dist: pandas>=1.3.0
-Requires-Dist: tqdm>=4.62.0
-Requires-Dist: joblib>=1.1.0
-Dynamic: author
-Dynamic: author-email
-Dynamic: classifier
-Dynamic: description
-Dynamic: description-content-type
-Dynamic: home-page
-Dynamic: keywords
-Dynamic: license-file
-Dynamic: requires-dist
-Dynamic: requires-python
-Dynamic: summary
-
-# Persian Sentiment Analyzer
-
-[![Python Version](https://img.shields.io/badge/python-3.6%2B-blue)](https://www.python.org/)
-[![License](https://img.shields.io/badge/license-MIT-green)](LICENSE)
-
-A Python library for sentiment analysis of Persian (Farsi) text, capable of classifying opinions as "recommended", "not_recommended", or "no_idea".
-
-## Features
-
-- **Text Preprocessing**: Normalization, tokenization, stemming, and stopword removal for Persian text
-- **Word Embeddings**: Built-in Word2Vec implementation for Persian language
-- **Sentiment Classification**: Logistic Regression classifier trained on Persian sentiment data
-- **Model Persistence**: Save and load trained models for future use
-- **Batch Processing**: Analyze sentiment for multiple texts at once
-
-## Installation
-
-```bash
-pip install persian-sentiment-analyzer
-```
-
-## Dependencies
-- Python 3.6+
-
-- hazm
-
-- gensim
-
-- scikit-learn
-
-- numpy
-
-- pandas
-
-## Usage
-
-### Basic Usage
-
-``` bash
-from persian_sentiment_analyzer import SentimentAnalyzer
-
-# Initialize with a pre-trained model
-analyzer = SentimentAnalyzer(model_path="path/to/pretrained_model")
-
-# Predict sentiment
-result = analyzer.predict("این محصول بسیار عالی است")
-print(result)  # Output: 'recommended'
-```
-
-### Training Your Own Model
-
-``` bash
-from persian_sentiment_analyzer import SentimentAnalyzer
-import pandas as pd
-
-# Load your dataset
-data = pd.read_csv("persian_reviews.csv")
-texts = data['text'].tolist()
-labels = data['label'].values  # 0: not_recommended, 1: recommended, 2: no_idea
-
-# Initialize analyzer
-analyzer = SentimentAnalyzer()
-
-# Preprocess and tokenize texts
-tokenized_texts = [analyzer.preprocessor.preprocess_text(text) for text in texts]
-
-# Train Word2Vec model
-analyzer.train_word2vec(tokenized_texts, vector_size=100)
-
-# Prepare feature vectors
-X = np.array([analyzer.sentence_vector(tokens) for tokens in tokenized_texts])
-
-# Train classifier
-analyzer.train_classifier(X, labels)
-
-# Save the trained model
-analyzer.save_model("my_persian_model")
-```
-
-### Batch Processing
-
-```bash
-from persian_sentiment_analyzer import predict_sentiments_for_file
-
-# Process a CSV file containing Persian comments
-results_summary = predict_sentiments_for_file(
-    analyzer,
-    input_file="comments.csv",
-    output_file="results.csv",
-    summary_file="summary.csv"
-)
-
-print(results_summary)
-```
-
-## Model Architecture
-1- Text Preprocessing:
-
-- Normalization (Hazm)
-
-- Tokenization
-
-- Stemming
-
-- Stopword removal
-
-2- Feature Extraction:
-
-- Word2Vec embeddings (100 dimensions)
-
-- Sentence vectors (average of word vectors)
-
-3- Classification:
-
-- Logistic Regression with L2 regularization
-
-## Performance
-The pre-trained model achieves the following performance on our test set:
-
-Metric	Value
-Accuracy	85.2%
-Precision	84.7%
-Recall	85.0%
-F1-score	84.8%
-
-License
-This project is licensed under the MIT [License](https://github.com/RezaGooner/Sentiment-Survey-Analyzer/blob/main/LICENSE) - see the LICENSE file for details
-
-
-> Github : RezaGooner

persent-0.1.6/PerSent.egg-info/top_level.txt

@@ -1 +0,0 @@
-PerSent

persent-0.1.6/README.md

@@ -1,130 +0,0 @@
-# Persian Sentiment Analyzer
-
-[![Python Version](https://img.shields.io/badge/python-3.6%2B-blue)](https://www.python.org/)
-[![License](https://img.shields.io/badge/license-MIT-green)](LICENSE)
-
-A Python library for sentiment analysis of Persian (Farsi) text, capable of classifying opinions as "recommended", "not_recommended", or "no_idea".
-
-## Features
-
-- **Text Preprocessing**: Normalization, tokenization, stemming, and stopword removal for Persian text
-- **Word Embeddings**: Built-in Word2Vec implementation for Persian language
-- **Sentiment Classification**: Logistic Regression classifier trained on Persian sentiment data
-- **Model Persistence**: Save and load trained models for future use
-- **Batch Processing**: Analyze sentiment for multiple texts at once
-
-## Installation
-
-```bash
-pip install persian-sentiment-analyzer
-```
-
-## Dependencies
-- Python 3.6+
-
-- hazm
-
-- gensim
-
-- scikit-learn
-
-- numpy
-
-- pandas
-
-## Usage
-
-### Basic Usage
-
-``` bash
-from persian_sentiment_analyzer import SentimentAnalyzer
-
-# Initialize with a pre-trained model
-analyzer = SentimentAnalyzer(model_path="path/to/pretrained_model")
-
-# Predict sentiment
-result = analyzer.predict("این محصول بسیار عالی است")
-print(result)  # Output: 'recommended'
-```
-
-### Training Your Own Model
-
-``` bash
-from persian_sentiment_analyzer import SentimentAnalyzer
-import pandas as pd
-
-# Load your dataset
-data = pd.read_csv("persian_reviews.csv")
-texts = data['text'].tolist()
-labels = data['label'].values  # 0: not_recommended, 1: recommended, 2: no_idea
-
-# Initialize analyzer
-analyzer = SentimentAnalyzer()
-
-# Preprocess and tokenize texts
-tokenized_texts = [analyzer.preprocessor.preprocess_text(text) for text in texts]
-
-# Train Word2Vec model
-analyzer.train_word2vec(tokenized_texts, vector_size=100)
-
-# Prepare feature vectors
-X = np.array([analyzer.sentence_vector(tokens) for tokens in tokenized_texts])
-
-# Train classifier
-analyzer.train_classifier(X, labels)
-
-# Save the trained model
-analyzer.save_model("my_persian_model")
-```
-
-### Batch Processing
-
-```bash
-from persian_sentiment_analyzer import predict_sentiments_for_file
-
-# Process a CSV file containing Persian comments
-results_summary = predict_sentiments_for_file(
-    analyzer,
-    input_file="comments.csv",
-    output_file="results.csv",
-    summary_file="summary.csv"
-)
-
-print(results_summary)
-```
-
-## Model Architecture
-1- Text Preprocessing:
-
-- Normalization (Hazm)
-
-- Tokenization
-
-- Stemming
-
-- Stopword removal
-
-2- Feature Extraction:
-
-- Word2Vec embeddings (100 dimensions)
-
-- Sentence vectors (average of word vectors)
-
-3- Classification:
-
-- Logistic Regression with L2 regularization
-
-## Performance
-The pre-trained model achieves the following performance on our test set:
-
-Metric	Value
-Accuracy	85.2%
-Precision	84.7%
-Recall	85.0%
-F1-score	84.8%
-
-License
-This project is licensed under the MIT [License](https://github.com/RezaGooner/Sentiment-Survey-Analyzer/blob/main/LICENSE) - see the LICENSE file for details
-
-
-> Github : RezaGooner