app.py

import gradio as gr
import pandas as pd
import joblib
from nltk.stem import WordNetLemmatizer
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.metrics import classification_report
from sklearn.compose import ColumnTransformer
from sklearn.impute import SimpleImputer
from textblob import TextBlob
import nltk
nltk.download('punkt')
nltk.download('wordnet')
import numpy as np

# Load pre-trained models and preprocessor objects
model = joblib.load('pet_adoption.pkl')
preprocessor = joblib.load('preprocessor.pkl')

# Lemmatizer for text preprocessing
lemmatizer = WordNetLemmatizer()

# Function to preprocess text input
def preprocess_text(text):
    text = text.lower()
    text = text.replace('[^\w\s]','',regex=True)
    text = text.replace('\s+',' ',regex=True)
    text = text.replace('\n',' ',regex=True)
    text = text.replace('\d+',' ',regex=True)
    stop_words = set(stopwords.words('english'))
    words = nltk.word_tokenize(text)
    words = [word for word in words if word not in stop_words]
 
    # Lemmatization
    words = [lemmatizer.lemmatize(word) for word in words]
    # Join processed words
    return ' '.join(words)


# Duygu analizi fonksiyonu
def analyze_sentiment(description):
    analysis = TextBlob(description)
    # Sentiment.polarity değeri -1 ile 1 arasında değişir: -1 negatif, 0 nötr, 1 pozitif
    return analysis.sentiment.polarity

def sentiment_to_adoption_speed(sentiment):
    if sentiment > 0.2:
        return 2  # Yüksek olumlu duygu
    elif sentiment < -0.2:
        return 0  # Yüksek olumsuz duygu
    else:
        return 1  # Nötr veya orta duygu



# Gradio arayüzü için girdi alanları ve işlemler
def predict_adoption_speed(age, breed1, breed2, gender, color1, color2, color3,
                           maturity_size, fur_length, vaccinated, dewormed,
                           sterilized, health, quantity, fee, state, video_amt,
                           photo_amt, description):
    
    # Convert input into DataFrame format
    data = pd.DataFrame({
        'Age': [age],
        'Breed1': [breed1],
        'Breed2': [breed2],
        'Gender': [gender],
        'Color1': [color1],
        'Color2': [color2],
        'Color3': [color3],
        'MaturitySize': [maturity_size],
        'FurLength': [fur_length],
        'Vaccinated': [vaccinated],
        'Dewormed': [dewormed],
        'Sterilized': [sterilized],
        'Health': [health],
        'Quantity': [quantity],
        'Fee': [fee],
        'State': [state],
        'VideoAmt': [video_amt],
        'PhotoAmt': [photo_amt],
        'sentiments': [sentiments]
    })
    
    # Separate numerical and categorical columns
    num_columns = ['Age', 'Quantity', 'Fee', 'VideoAmt', 'PhotoAmt']
    cat_columns = ['Breed1', 'Breed2', 'Gender', 'Color1', 'Color2', 'Color3',
                   'MaturitySize', 'FurLength', 'Vaccinated', 'Dewormed',
                   'Sterilized', 'Health', 'State']
    
    # Preprocess numerical and categorical data
    processed_data = preprocessor.transform(data)
    
    # Convert description text to sentiments
    cleaned_description = preprocess_text(description)
    sentiment = analyze_sentiment(cleaned_description)
    sentiments = sentiment_to_adoption_speed(sentiment)  # Placeholder for sentiment analysis
    
    # Make prediction
    prediction = model.predict(processed_data)
    
    # Return prediction result
    return int(prediction[0])

# Gradio arayüzünü oluşturma

# Define Gradio Interface
iface = gr.Interface(
    fn=predict_adoption_speed, 
    inputs=[
        gr.Slider(minimum=0, maximum=20, label="Age"),
        gr.Number(label="Breed1"),
        gr.Number(label="Breed2"),
        gr.Dropdown(choices=[0, 1, 2], label="Gender"),
        gr.Dropdown(choices=[1, 2, 3, 4, 5, 6, 7], label="Color1"),
        gr.Dropdown(choices=[0, 1, 2, 3, 4, 5, 6, 7], label="Color2"),
        gr.Dropdown(choices=[0, 1, 2, 3, 4, 5, 6, 7], label="Color3"),
        gr.Dropdown(choices=[0, 1, 2, 3], label="MaturitySize"),
        gr.Dropdown(choices=[1, 2, 3], label="FurLength"),
        gr.Dropdown(choices=[0, 1], label="Vaccinated"),
        gr.Dropdown(choices=[0, 1], label="Dewormed"),
        gr.Dropdown(choices=[0, 1], label="Sterilized"),
        gr.Dropdown(choices=[0, 1], label="Health"),
        gr.Number(label="Quantity"),
        gr.Number(label="Fee"),
        gr.Dropdown(choices=[41324, 41325, 41326, 41327, 41330, 41332, 41335, 41336, 41342, 41345, 41361, 41367, 41401, 41415, 41452], label="State"),
        gr.Slider(minimum=0, maximum=10, label="VideoAmt"),
        gr.Slider(minimum=0, maximum=10, label="PhotoAmt"),
        gr.Textbox(label="Description", placeholder="Enter pet description...")
    ],
    outputs=gr.Textbox(label="Predicted Adoption Speed")
)

# Launch Gradio Interface
iface.launch(share=True)