Upload 12 files

modelasdfasdfsad

sentimentAnalysis/app/model (2).pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb112aa1b00439676911e46781a363e89c53908be22c829844b8ad5ae2dbd5a4
+size 438041938

sentimentAnalysis/app/sentiment_analysis.py

@@ -0,0 +1,66 @@
+import joblib
+import re
+import os
+import nltk
+from nltk.tokenize import word_tokenize
+from nltk.corpus import stopwords
+from nltk.stem import PorterStemmer, WordNetLemmatizer
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.naive_bayes import MultinomialNB
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
+
+nltk.download('punkt_tab')
+nltk.download('stopwords')
+nltk.download('wordnet')
+
+
+# Get the directory of this file
+current_dir = os.path.dirname(os.path.abspath(__file__))
+
+# Load the trained model and vectorizer
+model_path = os.path.join(current_dir, "model (2).pkl")
+vectorizer_path = os.path.join(current_dir, "tokenizer (2).pkl")
+
+model = joblib.load(model_path)
+vectorizer = joblib.load(vectorizer_path)
+
+# Preprocessing function
+def preprocess_text(text, use_stemming=False, use_lemmatization=True):
+    text = text.lower()
+    text = re.sub(r'\W', ' ', text)
+    words = word_tokenize(text)
+    
+    stop_words = set(stopwords.words('english'))
+    stop_words.discard('not')  # Keep 'not' for sentiment analysis
+    words = [word for word in words if word not in stop_words]
+
+    stemmer = PorterStemmer()
+    lemmatizer = WordNetLemmatizer()
+
+    if use_stemming:
+        words = [stemmer.stem(word) for word in words]
+    elif use_lemmatization:
+        words = [lemmatizer.lemmatize(word) for word in words]
+
+    return " ".join(words)
+
+# Prediction function
+def predict_sentiment(analyser):
+    """Predicts sentiment using the trained BERT model."""
+    processed_text = preprocess_text(analyser.sentence)  # ✅ Preprocess the text
+
+    # ✅ Tokenize input text (Replacing vectorizer.transform)
+    inputs = vectorizer(processed_text, truncation=True, padding="max_length", max_length=256, return_tensors="pt")
+
+    # ✅ Move inputs to the correct device
+   #inputs = {key: val.to(device) for key, val in inputs.items()}
+
+    # ✅ Get model prediction
+    with torch.no_grad():
+        outputs = model(**inputs)
+        prediction = torch.argmax(outputs.logits, dim=1).item()
+
+    # ✅ Convert prediction to sentiment label
+    sentiment_labels = ["Negative", "Neutral", "Positive"]
+    return sentiment_labels[prediction]

sentimentAnalysis/app/service.py

@@ -0,0 +1,13 @@
+from fastapi import APIRouter
+from pydantic import BaseModel
+from app.sentiment_analysis import predict_sentiment  
+
+class SentimentAnalyser(BaseModel):
+    sentence: str
+
+router = APIRouter()
+
+@router.post("/sentiment-analyser/post/")
+async def create_grade(analyser: SentimentAnalyser):
+    result = predict_sentiment(analyser)
+    return {"sentiment": result}

sentimentAnalysis/app/tokenizer (2).pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43338c2179e9acd6b91517313d27f6de37937464fda7f47d11c1eb2c6134839c
+size 629587

sentimentAnalysis/main.py

@@ -0,0 +1,13 @@
+from fastapi import FastAPI
+from app.service import router as sentiment_analysis_router
+
+# Initialize FastAPI app
+app = FastAPI()
+
+# Include both APIs
+app.include_router(sentiment_analysis_router, prefix="/api", tags=["SentimentAnalyser"])
+
+@app.get("/")
+async def root():
+    return {"message": "Welcome to backend"}
+