Water_Potability_webapp.py

import streamlit as st
import pandas as pd
import matplotlib.pyplot as plt 
import numpy as np
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split, cross_val_score
#rom sklearn.metrics import confusion_matrix, classification_report
#from sklearn.metrics import precision_score, recall_score,f1_score
#from sklearn.metrics import plot_roc_curve

header = st.container()
dataset = st.container()
features = st.container()
inputs = st.container()

#@st.cache(for running it for one time if the file name is same then its not going to run again)
#def get_data():
#  taxi_data=pd.read_csv

# For customization u can use css:
# =============================================================================
# st.markdown(
#     """
#     <style>
#     .main {
#     background-color:#F5F5F5;
#     }
#     </style>
#     """,
#     unsafe_allow_html=True)
# =============================================================================

with header:
    st.title("Hydr8")
    st.text("In this project we will find out whether the water is potable or not")

with dataset:
    st.header("Water Potability Dataset")
    st.text("I found this data set on Kaggle")
    
    df=pd.read_csv(r"C:\Users\hp\Desktop\Water-Potability\water-potability11.csv")
    st.write(df.head())
    
    st.subheader("Potability of Water")
    a=df['Potability'].value_counts()
    st.bar_chart(a,height=500)


with features:
    st.header("The features on which the model is trained")
    st.markdown('* pH of Water')
    st.markdown('* Hardness of Water')
    st.markdown('* Solids in water')
    st.markdown('* Chloroamines in water')
    st.markdown('* Sulfate in water')
    st.markdown('* Conductivity of water')
    st.markdown('* Organic Carbon in water')
    st.markdown('* Trihalomethanes in water')
    st.markdown('* Turbidity in water')
    st.markdown('* Potability of water')
    
with inputs:
    st.header("Time to take input from the user")
    st.text("Enter the inputs according to the parameters of the water taken")
    
    sel_col,disp_col = st.columns(2)
    
    ph=sel_col.slider('pH',min_value=0,max_value=14,value=7)
    hardness=sel_col.selectbox("Hardness",options=[50,100,150,200,250],index=0)
    solids=sel_col.slider('Solids',min_value=10000,max_value=25000,value=15000)
    chloroamines=sel_col.slider('Chloroamines',min_value=1,max_value=14,value=7,step=1)
    sulfate=sel_col.slider('Sulfate',min_value=50,max_value=400,value=200)
    conductivity=sel_col.slider('Conductivity',min_value=100,max_value=500,value=300)
    organic_carbon=sel_col.slider('Organic Carbon',min_value=5,max_value=25,value=14)
    trihalomethanes=sel_col.slider('Trihalomethanes',min_value=20,max_value=100,value=50)
    turbidity=sel_col.slider('Turbidity',min_value=1,max_value=7,value=4)    
    #n_estimators=sel_col.slider('How many trees should be there', min_value=10,max_value=1000,value=200,step=10)
    arr=[[ph,hardness,solids,chloroamines,sulfate,conductivity,organic_carbon,trihalomethanes,turbidity]]
    #input_feature = sel_col.text_input("Which feature should be used as the input feature","pH")
    
    clf=RandomForestClassifier(max_depth=None,n_estimators=310,min_samples_leaf=5,min_samples_split=7)
    X=df.drop('Potability',axis=1)
    y=df['Potability']
    clf.fit(X,y)
    
    y_pred=clf.predict(arr)
    if y_pred==1:
        st.write("""### Water is Potable""")
    else:
        st.write("### Water is not Potable")