import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
import pickle
import gradio as gr
from math import sqrt
from datasets import load_dataset


df = load_dataset("csv", data_files = "vishal323/heart.csv")


df

df.info()

cp_data= df['cp'].value_counts().reset_index()
cp_data['index'][3]= 'asymptomatic'
cp_data['index'][2]= 'non-anginal'
cp_data['index'][1]= 'Atyppical Anigma'
cp_data['index'][0]= 'Typical Anigma'
cp_data

ecg_data= df['restecg'].value_counts().reset_index()
ecg_data['index'][0]= 'normal'
ecg_data['index'][1]= 'having ST-T wave abnormality'
ecg_data['index'][2]= 'showing probable or definite left ventricular hypertrophy by Estes'

ecg_data

def outbreak(feature):

  fig = plt.figure()
  plt.rcParams.update({'font.size': 10})
  plt.rc('xtick', labelsize=5)

  if (feature == "Age"):

    
    plt.title("Age of Patients")
    plt.xlabel("Age")


    sns.countplot(x='age',data=df);
    return fig

  elif (feature == "Sex"):
    plt.title("Sex of Patients,0=Female and 1=Male")
    sns.countplot(x='sex',data=df);
    return fig

  elif (feature == "Chest Pain"):
    plt.title("Chest Pain of Patients")

    sns.barplot(x=cp_data['index'],y= cp_data['cp']);
    return fig

  elif (feature == "ECG"):
    plt.title("ECG data of Patients")

    sns.barplot(x=ecg_data['index'],y= ecg_data['restecg']);
    return fig

  elif (feature == "Blood Pressure"):
    plt.title("Resting Blood Pressure (mmHg)")

    sns.distplot(df['trestbps'], kde=True, color = 'magenta')
    plt.xlabel("Resting Blood Pressure (mmHg)")
    return fig

def op(target, sex, cp, age, bp, ch):

  fig = plt.figure()
  plt.rcParams.update({'font.size': 10})
  plt.rc('xtick', labelsize=5)

  print(target, sex, cp, age, bp, ch)

  data = df[((df['target'] == 1) & df['sex'] == sex) & (df['cp'] == cp) & (df['age'] >= age) & (df['trestbps'] >= bp) & (df['chol'] >= ch)  ]
  
  if (data.empty):
    return fig

  if (target == "Age"):
    
    plt.title("Count of age of diseased people")
    plt.xlabel("Age")
    sns.countplot(x='age',data=data);
    return fig

  elif (target == "Sex"):
    
    plt.title("Count of sex of diseased people")
    plt.xlabel("Sex")
    sns.countplot(x='sex',data=data);
    return fig

  if (target == "Chest Pain"):
    
    plt.title("Count of diseased people with cheast pain")
    plt.xlabel("Chest Pain")
    sns.countplot(x='cp',data=data);
    return fig

  if (target == "ECG"):
    
    plt.title("Count of people with low glucose")
    plt.xlabel("ECG")
    sns.countplot(x='restecg',data=data);
    return fig

  if (target == "Blood Pressure"):
    
    plt.title("Count of diseased people with high BP")
    plt.xlabel("BP")
    sns.countplot(x='trestbps',data=data);
    return fig

def prd(model, age, sex,	cp,	trestbps,	chol,	fbs,	restecg,	thalach,	exang,	oldpeak,	slope,	ca,	thal):

  if model == "Random Forest":

    filename = 'DACVV/randomforest.pkl'
    X_test =  np.array([[age,	sex,	cp,	trestbps,	chol,	fbs,	restecg,	thalach,	exang,	oldpeak,	slope,	ca,	thal],[52,	1,	0,	125,	212,	0,	1,	168,	0,	1.0,	2,	2,	3]])
    loaded_model = pickle.load(open(filename, 'rb'))
    result = loaded_model.predict(X_test)[0]
  
  else:
    filename = 'DACVV/scaling.pkl'
    X_test =  np.array([[age,	sex,	cp,	trestbps,	chol,	fbs,	restecg,	thalach,	exang,	oldpeak,	slope,	ca,	thal],[52,	1,	0,	125,	212,	0,	1,	168,	0,	1.0,	2,	2,	3]])
    loaded_model = pickle.load(open(filename, 'rb'))
    result = loaded_model.predict(X_test)[0]


  return "😔, You may a Heart Disease" if (result == 1) else "😁, You are Healthy!!!"

inputs = gr.Dropdown(["Age", "Sex", "Chest Pain", "ECG", "Blood Pressure"], label="Input Feature")

outputs = gr.Plot()

visualisation = gr.Interface(
    fn=outbreak,
    inputs=inputs,
    outputs=outputs,
    
)

vis = gr.Interface(
    inputs = [
        gr.Radio(["Age", "Sex", "Chest Pain", "ECG", "Blood Pressure"], label = "Target Feature"),
        gr.Radio([1, 0], label = "Sex"),
        gr.Radio([0,1,2,3], label = "Chest Pain"),
        gr.Slider(25, 80, value=50, step = 1, label = "Age"),
        gr.Slider(94, 200, value=150, step = 1, label = "Blood Pressure"),
        gr.Slider(126, 564, value=130, step = 1, label = "Cholestrol")
        
    ],

    fn=op,
    outputs = gr.Plot(),

    
    examples=[
        ["Age", 1,  2, 50, 100, 222],
        ["Sex", 0, 1, 30, 150, 322],
        ["Chest Pain", 1, 0, 40, 120, 422],
        ["ECG", 1, 3, 70, 98, 522],
        ["Blood Pressure", 0, 1, 28, 170, 262],
    ]
)

pred = gr.Interface(
    inputs = [gr.Radio(["Random Forest", "Scaler"], label = "Model"),
              "number",
              gr.Radio([0, 1], label = "Sex"),
              gr.Radio([0,1,2,3], label = "Chest Pain"),
              gr.Slider(94, 200, value=150, step = 1, label = "Blood Pressure"),
              gr.Slider(126, 564, value=130, step = 1, label = "Cholestrol"),
              gr.Radio([0, 1], label = "FBS"),
              gr.Radio([0, 1, 2], label = "RestECG"),
              gr.Slider(71, 202, value=50, step = 1, label = "Thalach"),
              gr.Radio([0, 1], label = "exang"),
              gr.Slider(0, 6.2, value=3, label = "OldPeak"),
              gr.Radio([0, 1, 2], label = "Slope"),
              gr.Slider(0, 4, value=3, step = 1, label = "CA"),
              gr.Slider(0, 3, value=50, step = 1, label = "Thal"),
              
              
              
              ],

    fn=prd,
    outputs = "text",

    
    examples=[
        ["Random Forest", 52,	1,	0,	125,	212,	0,	1,	168,	0,	1.0,	2,	2,	3],
        ["Scaler", 62,	0,	0,	138,	294,	1,	1,	106,	0,	1.9,	1,	3,	2],
        ["Random Forest", 44,	0,	2,	108,	141,	0,	1,	175,	0,	0.6,	1,	0,	2],
        ["Scaler", 68,	0,	2,	120,	211,	0,	0,	115,	0,	1.5,	1,	0,	2]
    ]
)

interface = gr.TabbedInterface([visualisation, vis, pred], ["Visualisation", "Real Time Analysis", "Predictions"])

interface.launch(inline = False)