import pandas as pd
import re
import gradio as gr
from sklearn.model_selection import train_test_split
from sklearn.pipeline import make_pipeline
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score
from collections import Counter

# 1. Load and clean data
df = pd.read_csv("spam.csv", encoding="latin1")[["v1", "v2"]]
df.columns = ["label", "text"]
df["label"] = df["label"].map({"ham": 0, "spam": 1})

# 2. Clean text
def clean_text(text):
    text = text.lower()
    text = re.sub(r"\W+", " ", text)
    return text.strip()

df["text"] = df["text"].apply(clean_text)

# 3. Split data
X_train, X_test, y_train, y_test = train_test_split(
    df["text"], df["label"], test_size=0.2, stratify=df["label"], random_state=42
)

# 4. Build and train model
model = make_pipeline(
    TfidfVectorizer(ngram_range=(1, 2), stop_words="english"),
    LogisticRegression(max_iter=1000, class_weight="balanced")
)

model.fit(X_train, y_train)

# 5. Evaluate
accuracy = accuracy_score(y_test, model.predict(X_test))
print(f"Validation Accuracy: {accuracy:.2%}")

# 6. Gradio prediction function
def predict_spam(message):
    cleaned = clean_text(message)
    pred = model.predict([cleaned])[0]
    prob = model.predict_proba([cleaned])[0][pred]
    label = "🚫 Spam" if pred == 1 else "📩 Not Spam (Ham)"
    return f"{label} (Confidence: {prob:.2%})"

# 7. Gradio UI
gr.Interface(
    fn=predict_spam,
    inputs=gr.Textbox(lines=4, label="Enter SMS Message"),
    outputs=gr.Text(label="Prediction"),
    title="SMS Spam Detector",
    description=f"Detects spam in SMS messages. Trained on uploaded CSV (Accuracy: {accuracy:.2%})."
).launch()