app.py

## CSCI4750/5750-SLU
# Jie Hou, Oct 5, 2022

import gradio as gr
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
import matplotlib.colors as colors
import itertools
from scipy.stats import norm
from scipy import stats
from sklearn.naive_bayes import GaussianNB



def gaussian(x, n, u, s):
    #u = x.mean()
    #s = x.std()

    # divide [x.min(), x.max()] by n
    x = np.linspace(x.min(), x.max(), n)

    a = ((x - u) ** 2) / (2 * (s ** 2))
    y = 1 / (s * np.sqrt(2 * np.pi)) * np.exp(-a)

    return x, y, u, s
    
def plot_figure(N, u1, std1, u2, std2, show_dist, classifier=None):
  #N = 100
  import numpy as np
  import matplotlib.pyplot as pp
  pp.style.use('default')
  val = 0. # this is the value where you want the data to appear on the y-axis.

  points_class1 = [stats.norm.rvs(loc=u1, scale = std1) for x in range(N)]
  points_class2 = [stats.norm.rvs(loc=u2, scale = std2) for x in range(N)]

  pd_class1 = pd.DataFrame({'Feature 1 (X)': points_class1, 'Label (Y)': np.repeat(0,len(points_class1))})
  pd_class2 = pd.DataFrame({'Feature 1 (X)': points_class2, 'Label (Y)': np.repeat(1,len(points_class2))})


  pd_all = pd.concat([pd_class1, pd_class2]).reset_index(drop=True)

  import numpy as np
  X_data= pd_all['Feature 1 (X)'].to_numpy().reshape((len(pd_all),1))
  y_labels= pd_all['Label (Y)']


  # define x, y limits
  x_min, x_max = X_data[:, 0].min() - 1, X_data[:, 0].max() + 1
  y_min, y_max = 0-1, 1 + 1

  fig = pp.figure(figsize=(8, 6)) # figure size in inches
  fig.subplots_adjust(left=0, right=1, bottom=0, top=1, hspace=0.3, wspace=0.05) 


  pp.tick_params(left = False, right = False , labelleft = False ,
                  labelbottom = True, bottom = False)

  #reference = [stats.uniform.rvs(loc=1, scale = 1) for x in range(N)]
  pp.plot(points_class1, np.zeros_like(points_class1) + val, 'x', label = 'Class 1', markersize = 10)
  pp.plot(points_class2, np.zeros_like(points_class2) + val, 'o', label = 'Class 2', markersize = 10)


  if show_dist:
    x = np.arange(x_min, x_max, 0.01) # define range of x
    x, y, u, s = gaussian(x, 10000, np.mean(points_class1), np.std(points_class1) )
    pp.plot(x, y)
    #pp.plot(x, y, label=r'$Gaussian (\mu=%.2f,\ \sigma=%.2f)$' % (u, s))


    x = np.arange(x_min, x_max, 0.01) # define range of x
    x, y, u, s = gaussian(x, 10000, np.mean(points_class2), np.std(points_class2) )
    pp.plot(x, y)
    #pp.plot(x, y, label=r'$Gaussian (\mu=%.2f,\ \sigma=%.2f)$' % (u, s))



  ### draw decision boundary on knn 

  import numpy as np
  from matplotlib import pyplot as plt
  from sklearn import neighbors, datasets
  from matplotlib.colors import ListedColormap

  # Create color maps for 3-class classification problem, as with iris
  cmap_light = ListedColormap(['#FFAAAA', '#AAFFAA'])
  cmap_bold = ListedColormap(['#FF0000', '#00FF00'])


  xx, yy = np.meshgrid(np.linspace(x_min, x_max, 100),
                          np.linspace(y_min, y_max, 100))

  
  if classifier == 'LDA':
    from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
    model_sk = LinearDiscriminantAnalysis()
    model_sk.fit(X_data,y_labels)
    zz = model_sk.predict(np.c_[xx.ravel()])

    #Predictions for each point on meshgrid
    #zz = np.array(  [model_sk.predict( [[xx,yy]])[0] for xx, yy in zip(np.ravel(X), np.ravel(Y)) ] )

    #Reshaping the predicted class into the meshgrid shape
    Z = zz.reshape(xx.shape)

    pp.pcolormesh(xx, yy, Z, cmap=cmap_light, alpha=0.2)
  elif classifier == 'QDA':
    from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis
    model_sk = QuadraticDiscriminantAnalysis()
    model_sk.fit(X_data,y_labels)
    
    model_sk.fit(X_data,y_labels)
    zz = model_sk.predict(np.c_[xx.ravel()])

    #Predictions for each point on meshgrid
    #zz = np.array(  [model_sk.predict( [[xx,yy]])[0] for xx, yy in zip(np.ravel(X), np.ravel(Y)) ] )

    #Reshaping the predicted class into the meshgrid shape
    Z = zz.reshape(xx.shape)
    
    print("Z: ",Z)
    pp.pcolormesh(xx, yy, Z, cmap=cmap_light, alpha=0.2)
  elif classifier == 'NaiveBayes':
    from sklearn.naive_bayes import GaussianNB
    model_sk = GaussianNB(priors = None)
    model_sk.fit(X_data,y_labels)
    Z = model_sk.predict(np.c_[xx.ravel()])
    Z = Z.reshape(xx.shape)

    pp.pcolormesh(xx, yy, Z, cmap=cmap_light, alpha=0.2)

  

  pp.xlim([x_min, x_max])
  pp.ylim([y_min, y_max])
  pp.xlabel("Feature 1 (X1)", size=20)
  pp.xticks(fontsize=20)
  pp.ylabel("Feature 2 (X2)")
  pp.legend(loc='upper right', borderpad=0, handletextpad=0, fontsize = 20)
  pp.savefig('plot.png')

  return 'plot.png', pd_all



set_mean_class1 = gr.Slider(-20, 20, step=0.5, default=1, label = 'Mean (Class 1)')
set_std_class1 = gr.Slider(0, 10, step=0.5, default=1.5, label = 'Standard Deviation (Class 1)')

# 2. define mean and standard deviation for class 2

set_mean_class2 = gr.Slider(-20, 20, step=0.5, default=10, label = 'Mean (Class 2)')
set_std_class2 = gr.Slider(0, 10, step=0.5, default=1.5, label = 'Standard Deviation (Class 2)')

# 3. Define the number of data points 
set_number_points = gr.Slider(10, 100, step=5, default=20, label = 'Number of samples in each class')

# 4. show distribution or not
set_show_dist = gr.Checkbox(label="Show data distribution")

# 5. set classifier type
set_classifier = gr.Dropdown(["None", "LDA", "QDA", "NaiveBayes"])

# 6. define output imagem model
set_out_plot_images = gr.Image(label="Data visualization")

set_out_plot_table = gr.Dataframe(type='pandas', label ='Simulated Dataset')





### configure gradio, detailed can be found at https://www.gradio.app/docs/#i_slider
interface = gr.Interface(fn=plot_figure, 
                         inputs=[set_number_points,set_mean_class1,set_std_class1,set_mean_class2,set_std_class2, set_show_dist, set_classifier], 
                         outputs=[set_out_plot_images,set_out_plot_table],
                         examples_per_page = 2,
                         #examples = get_sample_data(10), 
                         title="CSCI4750/5750 Demo: Web Application for Probabilistic Classifier (Single feature)", 
                         description= "Click examples below for a quick demo",
                         theme = 'huggingface',
                         layout = 'vertical'
                         )
interface.launch(debug=True)