SHARE
TWEET

Untitled

a guest Oct 21st, 2019 101 Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
  1. import streamlit as st
  2. from datetime import time
  3. from datetime import date
  4. import pandas as pd
  5. import matplotlib.pyplot as plt
  6. import seaborn as sns
  7. import plotly.figure_factory as ff
  8. import plotly.graph_objs  as go
  9. from sklearn.linear_model import LinearRegression
  10. from sklearn.model_selection import train_test_split
  11. from sklearn.metrics import mean_squared_error, r2_score
  12. from math import sqrt
  13.  
  14. import numpy as np
  15. sns.set_style("darkgrid")
  16.  
  17.  
  18. st.title("Advertisement and Sales Data")
  19.  
  20. st.markdown("""
  21.     The data set contains information about money spent on advertisement and their generated sales. Money
  22.     was spent on TV, radio and newspaper ads.
  23.  
  24.     ## Problem Statement
  25.     Sales (in thousands of units) for a particular product as a function of advertising budgets (in thousands of
  26.     dollars) for TV, radio, and newspaper media. Suppose that in our role as statistical consultants we are
  27.     asked to suggest.
  28.  
  29.     Here are a few important questions that you might seek to address:
  30.     - Is there a relationship between advertising budget and sales?
  31.     - How strong is the relationship between the advertising budget and sales?
  32.     - Which media contribute to sales?
  33.     - How accurately can we estimate the effect of each medium on sales?
  34.     - How accurately can we predict future sales?
  35.     - Is the relationship linear?
  36.  
  37.     We want to find a function that given input budgets for TV, radio and newspaper predicts the output sales
  38.     and visualize the relationship between the features and the response using scatter plots.
  39.  
  40.     The objective is to use linear regression to understand how advertisement spending impacts sales.
  41.    
  42.     ### Data Description
  43.     TV
  44.     Radio
  45.     Newspaper
  46.     Sales
  47.  
  48.  
  49. """)
  50. st.sidebar.title("Operations on the Dataset")
  51.  
  52. #st.subheader("Checkbox")
  53. w1 = st.sidebar.checkbox("show table", False)
  54. plot= st.sidebar.checkbox("show plots", False)
  55. plothist= st.sidebar.checkbox("show hist plots", False)
  56. trainmodel= st.sidebar.checkbox("Train model", False)
  57. dokfold= st.sidebar.checkbox("DO KFold", False)
  58. distView=st.sidebar.checkbox("Dist View", False)
  59. _3dplot=st.sidebar.checkbox("3D plots", False)
  60. linechart=st.sidebar.checkbox("Linechart",False)
  61. #st.write(w1)
  62.  
  63.  
  64. @st.cache
  65. def read_data():
  66.     return pd.read_csv("../ML_Python_LinearR_Case_2_Advertising_Data.csv")[["TV","radio","newspaper","sales"]]
  67.  
  68. df=read_data()
  69.  
  70. #st.write(df)
  71. if w1:
  72.     st.dataframe(df,width=2000,height=500)
  73. if linechart:
  74.     st.subheader("Line chart")
  75.     st.line_chart(df)
  76. if plothist:
  77.     st.subheader("Distributions of each columns")
  78.     options = ("TV","radio","newspaper","sales")
  79.     sel_cols = st.selectbox("select columns", options,1)
  80.     st.write(sel_cols)
  81.     #f=plt.figure()
  82.     fig = go.Histogram(x=df[sel_cols],nbinsx=50)
  83.     st.plotly_chart([fig])
  84.    
  85.  
  86. #    plt.hist(df[sel_cols])
  87. #    plt.xlabel(sel_cols)
  88. #    plt.ylabel("sales")
  89. #    plt.title(f"{sel_cols} vs Sales")
  90.     #plt.show()
  91. #    st.plotly_chart(f)
  92.  
  93. if plot:
  94.     st.subheader("correlation between sales and Ad compaigns")
  95.     options = ("TV","radio","newspaper","sales")
  96.     w7 = st.selectbox("Ad medium", options,1)
  97.     st.write(w7)
  98.     f=plt.figure()
  99.     plt.scatter(df[w7],df["sales"])
  100.     plt.xlabel(w7)
  101.     plt.ylabel("sales")
  102.     plt.title(f"{w7} vs Sales")
  103.     #plt.show()
  104.     st.plotly_chart(f)
  105.  
  106.  
  107. if distView:
  108.     st.subheader("Combined distribution viewer")
  109.     # Add histogram data
  110.  
  111.     # Group data together
  112.     hist_data = [df["TV"].values,df["radio"].values,df["newspaper"].values]
  113.  
  114.     group_labels = ["TV", "Radio", "newspaper"]
  115.  
  116.     # Create distplot with custom bin_size
  117.     fig = ff.create_distplot(hist_data, group_labels, bin_size=[0.1, 0.25, 0.5])
  118.  
  119.     # Plot!
  120.     st.plotly_chart(fig)
  121.  
  122. if _3dplot:
  123.     options = st.multiselect(
  124.      'Enter columns to plot',('TV', 'radio'),('TV', 'radio', 'newspaper', 'sales'))
  125.     st.write('You selected:', options)
  126.     st.subheader("TV & Radio vs Sales")
  127.     hist_data = [df["TV"].values,df["radio"].values,df["newspaper"].values]
  128.  
  129.     #x, y, z = np.random.multivariate_normal(np.array([0, 0, 0]), np.eye(3), 400).transpose()
  130.     trace1 = go.Scatter3d(
  131.         x=hist_data[0],
  132.         y=hist_data[1],
  133.         z=df["sales"].values,
  134.         mode="markers",
  135.         marker=dict(
  136.             size=8,
  137.             #color=df['sales'],  # set color to an array/list of desired values
  138.             colorscale="Viridis",  # choose a colorscale
  139.     #        opacity=0.,
  140.         ),
  141.     )
  142.  
  143.     data = [trace1]
  144.     layout = go.Layout(margin=dict(l=0, r=0, b=0, t=0))
  145.     fig = go.Figure(data=data, layout=layout)
  146.     st.write(fig)
  147.  
  148.  
  149.  
  150. # trainmodel= st.checkbox("Train model", False)
  151.  
  152. if trainmodel:
  153.     st.header("Modeling")
  154.     y=df.sales
  155.     X=df[["TV","radio","newspaper"]].values
  156.     X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3)
  157.  
  158.     lrgr = LinearRegression()
  159.     lrgr.fit(X_train,y_train)
  160.     pred = lrgr.predict(X_test)
  161.  
  162.     mse = mean_squared_error(y_test,pred)
  163.     rmse = sqrt(mse)
  164.  
  165.     st.markdown(f"""
  166.  
  167.     Linear Regression model trained :
  168.         - MSE:{mse}
  169.         - RMSE:{rmse}
  170.     """)
  171.     st.success('Model trained successfully')
  172.  
  173.  
  174. if dokfold:
  175.     st.subheader("KFOLD Random sampling Evalution")
  176.     st.empty()
  177.     my_bar = st.progress(0)
  178.  
  179.     from sklearn.model_selection import KFold
  180.  
  181.     X=df.values[:,-1].reshape(-1,1)
  182.     y=df.values[:,-1]
  183.     #st.progress()
  184.     kf=KFold(n_splits=10)
  185.     #X=X.reshape(-1,1)
  186.     mse_list=[]
  187.     rmse_list=[]
  188.     r2_list=[]
  189.     idx=1
  190.     fig=plt.figure()
  191.     i=0
  192.     for train_index, test_index in kf.split(X):
  193.     #   st.progress()
  194.         my_bar.progress(idx*10)
  195.         X_train, X_test = X[train_index], X[test_index]
  196.         y_train, y_test = y[train_index], y[test_index]
  197.         lrgr = LinearRegression()
  198.         lrgr.fit(X_train,y_train)
  199.         pred = lrgr.predict(X_test)
  200.        
  201.         mse = mean_squared_error(y_test,pred)
  202.         rmse = sqrt(mse)
  203.         r2=r2_score(y_test,pred)
  204.         mse_list.append(mse)
  205.         rmse_list.append(rmse)
  206.         r2_list.append(r2)
  207.         plt.plot(pred,label=f"dataset-{idx}")
  208.         idx+=1
  209.     plt.legend()
  210.     plt.xlabel("Data points")
  211.     plt.ylabel("PRedictions")
  212.     plt.show()
  213.     st.plotly_chart(fig)
  214.  
  215.     res=pd.DataFrame(columns=["MSE","RMSE","r2_SCORE"])
  216.     res["MSE"]=mse_list
  217.     res["RMSE"]=rmse_list
  218.     res["r2_SCORE"]=r2_list
  219.  
  220.     st.write(res)
  221.     st.balloons()
  222. #st.subheader("results of KFOLD")
  223.  
  224. #f=res.plot(kind='box',subplots=True)
  225. #st.plotly_chart([f])
RAW Paste Data
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy. OK, I Understand
 
Top