Upload 14 files

Dockerfile

@@ -0,0 +1,30 @@
+# Use Python 3.12 full image (not slim)
+FROM python:3.12
+
+# Set working directory
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    build-essential \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements first for better caching
+COPY requirements.txt .
+
+# Install Python dependencies
+RUN pip install --no-cache-dir --upgrade pip && \
+    pip install --no-cache-dir -r requirements.txt
+
+# Copy application code and models
+COPY . .
+
+# Expose port 7860 (Hugging Face Spaces default)
+EXPOSE 7860
+
+# Set environment variables for Hugging Face
+ENV GRADIO_SERVER_NAME="0.0.0.0"
+ENV GRADIO_SERVER_PORT=7860
+
+# Run the application
+CMD ["python", "app.py"]

app.py

@@ -0,0 +1,202 @@
+#!/usr/bin/env python3
+"""
+MEDIGUARD API - FastAPI Application
+Running on HuggingFace Spaces
+"""
+
+import numpy as np
+import pandas as pd
+import joblib
+from pathlib import Path
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel, Field
+from typing import List, Dict
+
+# ============================================================
+# CONFIGURATION
+# ============================================================
+
+MODEL_DIR = Path("models")
+
+CLINICAL_RANGES = {
+    "Glucose": (60, 300),
+    "Cholesterol": (100, 350),
+    "Hemoglobin": (8, 20),
+    "Platelets": (50000, 700000),
+    "White Blood Cells": (3000, 20000),
+    "Red Blood Cells": (3.0, 7.0),
+    "Hematocrit": (25, 60),
+    "Mean Corpuscular Volume": (65, 110),
+    "Mean Corpuscular Hemoglobin": (20, 40),
+    "Mean Corpuscular Hemoglobin Concentration": (28, 38),
+    "Insulin": (1, 60),
+    "BMI": (12, 50),
+    "Systolic Blood Pressure": (80, 200),
+    "Diastolic Blood Pressure": (40, 130),
+    "Triglycerides": (40, 600),
+    "HbA1c": (3.0, 14.0),
+    "LDL Cholesterol": (40, 250),
+    "HDL Cholesterol": (10, 120),
+    "ALT": (5, 120),
+    "AST": (5, 120),
+    "Heart Rate": (40, 220),
+    "Creatinine": (0.2, 5.0),
+    "Troponin": (0, 10),
+    "C-reactive Protein": (0, 100),
+}
+
+FEATURE_ORDER = list(CLINICAL_RANGES.keys())
+
+# ============================================================
+# LOAD MODELS AT STARTUP
+# ============================================================
+
+print("🏥 Loading MediGuard models...")
+
+try:
+    le = joblib.load(MODEL_DIR / "label_encoder.pkl")
+    scaler = joblib.load(MODEL_DIR / "scaler.pkl")
+    features_list = joblib.load(MODEL_DIR / "features.pkl")
+    meta = joblib.load(MODEL_DIR / "meta_neural.pkl")
+    metadata = joblib.load(MODEL_DIR / "metadata.pkl")
+    
+    base_models = []
+    for f in sorted(MODEL_DIR.glob("base_*.pkl")):
+        model = joblib.load(f)
+        name = f.stem.replace("base_", "")
+        base_models.append((name, model))
+    
+    MODELS_LOADED = True
+    print(f"✓ Loaded {len(base_models)} base models")
+    print(f"✓ Features: {len(features_list)}")
+    print(f"✓ Classes: {len(le.classes_)}")
+    
+except Exception as e:
+    MODELS_LOADED = False
+    print(f"❌ Error loading models: {e}")
+
+# ============================================================
+# PYDANTIC MODELS
+# ============================================================
+
+class PatientInput(BaseModel):
+    Glucose: float
+    Cholesterol: float
+    Hemoglobin: float
+    Platelets: float
+    White_Blood_Cells: float = Field(..., alias="White Blood Cells")
+    Red_Blood_Cells: float = Field(..., alias="Red Blood Cells")
+    Hematocrit: float
+    Mean_Corpuscular_Volume: float = Field(..., alias="Mean Corpuscular Volume")
+    Mean_Corpuscular_Hemoglobin: float = Field(..., alias="Mean Corpuscular Hemoglobin")
+    Mean_Corpuscular_Hemoglobin_Concentration: float = Field(..., alias="Mean Corpuscular Hemoglobin Concentration")
+    Insulin: float
+    BMI: float
+    Systolic_Blood_Pressure: float = Field(..., alias="Systolic Blood Pressure")
+    Diastolic_Blood_Pressure: float = Field(..., alias="Diastolic Blood Pressure")
+    Triglycerides: float
+    HbA1c: float
+    LDL_Cholesterol: float = Field(..., alias="LDL Cholesterol")
+    HDL_Cholesterol: float = Field(..., alias="HDL Cholesterol")
+    ALT: float
+    AST: float
+    Heart_Rate: float = Field(..., alias="Heart Rate")
+    Creatinine: float
+    Troponin: float
+    C_reactive_Protein: float = Field(..., alias="C-reactive Protein")
+
+    class Config:
+        populate_by_name = True
+
+
+class PredictionResult(BaseModel):
+    prediction: str
+    confidence: float
+    top_5_predictions: List[Dict[str, float]]
+    scaled_values: Dict[str, float]
+    model_version: str = "1.0"
+
+
+class HealthResponse(BaseModel):
+    status: str
+    models_loaded: bool
+    n_base_models: int
+    n_diseases: int
+
+
+# ============================================================
+# HELPER FUNCTIONS
+# ============================================================
+
+def scale_value(value: float, feature: str) -> float:
+    mn, mx = CLINICAL_RANGES[feature]
+    clipped = max(mn, min(mx, float(value)))
+    return (clipped - mn) / (mx - mn)
+
+
+def engineer_features(input_df: pd.DataFrame) -> pd.DataFrame:
+    df = input_df.copy()
+    # NOTE: If you want the FULL feature engineering block here,
+    # paste it EXACTLY from your original code.
+    return df
+
+
+def predict_disease(raw_values: Dict[str, float]) -> Dict:
+    scaled_values = {f: scale_value(v, f) for f, v in raw_values.items()}
+    input_df = pd.DataFrame([scaled_values])
+    input_engineered = engineer_features(input_df)
+
+    for feat in features_list:
+        if feat not in input_engineered.columns:
+            input_engineered[feat] = 0
+        
+    input_engineered = input_engineered[features_list]
+    X = scaler.transform(input_engineered.values)
+
+    base_probs = [model.predict_proba(X) for _, model in base_models]
+    meta_features = np.hstack(base_probs)
+    probs = meta.predict_proba(meta_features)[0]
+
+    pred_idx = np.argmax(probs)
+    disease = le.inverse_transform([pred_idx])[0]
+    confidence = float(probs[pred_idx])
+
+    top5_idx = np.argsort(probs)[-5:][::-1]
+    top5 = [{"disease": le.inverse_transform([i])[0],
+             "probability": float(probs[i])} for i in top5_idx]
+
+    return {
+        "prediction": disease,
+        "confidence": confidence,
+        "top_5_predictions": top5,
+        "scaled_values": scaled_values
+    }
+
+# ============================================================
+# FASTAPI BACKEND
+# ============================================================
+
+app = FastAPI(title="MediGuard API Backend")
+
+@app.get("/health")
+def health():
+    return {
+        "status": "healthy" if MODELS_LOADED else "unhealthy",
+        "models_loaded": MODELS_LOADED,
+        "n_base_models": len(base_models),
+        "n_diseases": len(le.classes_),
+    }
+
+@app.post("/predict", response_model=PredictionResult)
+def predict(patient: PatientInput):
+    raw_values = {k: getattr(patient, k.replace(" ", "_")) for k in CLINICAL_RANGES.keys()}
+    result = predict_disease(raw_values)
+    return PredictionResult(**result)
+
+# ============================================================
+# RUN THE APPLICATION
+# ============================================================
+
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=7860)

models/base_LightGBM.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:063f7957f26bc43bcacc3c8fe166e439570c8d53b089485622de6dbeb610a200
+size 10054996

models/base_XGB_Aggressive.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a90ee13044e8ddace99515840c094f8458282f88879b019363f1995894392e7
+size 5631588

models/base_XGB_Balanced.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ae71f7735d5a7c412788f9cf0d9d72bade14fdf1f99f2b6a5b81a0f4a4a73b2
+size 7123421

models/base_XGB_Conservative.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b9395cc9568629b3cbaaa11bc880f2546c72b413bb07381a7f8e64cfbc92e72
+size 6701906

models/base_XGB_VeryDeep.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb2a9712e3a688c134c63988be58815355c805c3efc77992114febc3c5582ebb
+size 6606718

models/base_XGB_Wide.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f97279d067528472b0a0251a349fc51181b56094787ad449b511ccbe9027e753
+size 7627616

models/features.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68b5838b7106e2180386d487ff1bda96eed9aec77e6d511c2700f461fab89961
+size 1209

models/label_encoder.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1bc1da9ba8b7a5b0b4c96f50fd1738fb605026c52feab7cf05a84da670b76a94
+size 774

models/meta_neural.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09bafaf8f9475ba8b790561a713eda070f375c0773040d6836af807faec8a926
+size 887853

models/metadata.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96438af3d822f20a48f27a3a3efb5ef2e074e31a4be6c979b76eacaeb81adda1
+size 554

models/scaler.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:287c4d983cde39b5eaac45b2a0e524593f38ebe1ba90a6ee1ebdad7334fe5f08
+size 2199

requirements.txt

@@ -0,0 +1,9 @@
+fastapi
+pydantic
+numpy
+pandas
+scikit-learn
+joblib
+xgboost
+lightgbm
+uvicorn