Merge branch 'master' of https://github.com/abheeeenav/DataMetric

Author: Ritavdas

__pycache__/app.cpython-38.pyc

@@ -3,7 +3,15 @@
 from authlib.integrations.flask_client import OAuth
 import os
 from datetime import timedelta
+#from forms import ProfileForm
 import pandas as pd
+#from forms import ProfileForm
+#from kmeans import K_cluster
+import io 
+import base64
+
+from werkzeug.utils import send_file
+from matplotlib.backends.backend_agg import FigureCanvasAgg as FigureCanvas
 import numpy as np
 import matplotlib.pyplot as plt
 from sklearn.metrics import silhouette_score
@@ -60,9 +68,19 @@ def cluster(df, col):
     plt.xlabel(col[0])
     plt.ylabel(col[1])
     ax.set_zlabel(col[2])
-    plt.show()
-    image_path = os.path('image'+'.png')
-    plt.savefig(image_path)
+    canvas = FigureCanvas(fig)
+    img = io.BytesIO()
+    plt.savefig('static/img.png')
+    return 'static/img.png'
+    
+    '''
+    plt.savefig(img)
+    img.seek(0)
+    return send_file(img, mimetype='img/png')
+    
+    '''
+    
+    
 
 def silhoutte(data):
     km_silhouette = []
@@ -75,6 +93,7 @@ def silhoutte(data):
     cluster_n = int(km_silhouette.index(max(km_silhouette)))+1
     return cluster_n
 
+
 @app.route('/')
 def index():
     return render_template('login.html')
@@ -86,10 +105,20 @@ def home():
     email = dict(session)['profile']['email']
     if request.method == 'POST':
         file = pd.read_csv(request.files['file'])
-        col =['CustomerID','Age','Income']
-        cluster(file,col)
+        col =['Age','Income','Score']
+        image = cluster(file,col)
+        return render_template('image.html', image = image)
     return render_template('index.html')
 
+
+'''
+@app.route('/profile')
+@login_required
+def profile():
+    form = ProfileForm()
+    return render_template('profile.html',title = 'Register',form=form)
+'''
+
 @app.route('/login')
 def login():
     google = oauth.create_client('google')  # create the google oauth client
@@ -99,18 +128,15 @@ def login():
 
 @app.route('/authorize')
 def authorize():
-    google = oauth.create_client('google')  # create the google oauth client
-    token = google.authorize_access_token()  # Access token from google (needed to get user info)
-    resp = google.get('userinfo')  # userinfo contains stuff u specificed in the scrope
+    google = oauth.create_client('google')
+    token = google.authorize_access_token()
+    resp = google.get('userinfo')  
     user_info = resp.json()
-    user = oauth.google.userinfo()  # uses openid endpoint to fetch user info
-    # Here you use the profile/user data that you got and query your database find/register the user
-    # and set ur own data in the session not the profile from google
+    user = oauth.google.userinfo()  
     session['profile'] = user_info
-    session.permanent = True  # make the session permanant so it keeps existing after broweser gets closed
+    session.permanent = True 
     return redirect('/home')
 
-
 @app.route('/logout')
 def logout():
     for key in list(session.keys()):

__pycache__/app.cpython-39.pyc

@@ -0,0 +1,10 @@
+from flask_wtf import FlaskForm
+from wtforms import StringField, SubmitField 
+from wtforms.validators import DataRequired, Length, Email
+
+class ProfileForm(FlaskForm):
+    name = StringField('Name', validators=[DataRequired(),Length(min =2,max=20)])
+    email = StringField('Email',validators =[DataRequired(),Email()])
+    company = StringField('Company',validators =[DataRequired()])
+    location = StringField('Location')
+    submit = SubmitField('Sign Up')

__pycache__/forms.cpython-39.pyc

@@ -1,3 +1,4 @@
+from app import cluster
 import numpy as np
 import pandas as pd
 import matplotlib.pyplot as plt
@@ -8,35 +9,42 @@
 import numpy as np
 import pandas as pd
 
-def cluster(df, col):
-    data = pd.DataFrame()
-    for j in col:
-        data[j]=df[j]
-    cluster_n =silhoutte(data)
-    km = KMeans(n_clusters=cluster_n)
-    clusters = km.fit_predict(data)
-    df["label"] = clusters
-    fig = plt.figure(figsize=(20,10))
-    ax = fig.add_subplot(111, projection='3d')
-    colors = ['blue','red','yellow','orange','green','purple']
-    count = 0
-    for i in range(cluster_n):
-        ax.scatter(df[col[0]][df.label == i], df[col[1]][df.label == i], df[col[2]][df.label == i], c=colors[count], s=60)
-        count+=1
-    ax.view_init(30, 185)
-    plt.xlabel(col[0])
-    plt.ylabel(col[1])
-    ax.set_zlabel(col[2])
-    plt.show()
+class K_cluster():
+    
+    def __init__(self, df, cluster_n, col):
+        self.df = df
+        self.col = col
+        self.cluster_n = cluster_n
 
-def silhoutte(data):
-    km_silhouette = []
-    for k in range(2,11):
-        kmeans = KMeans(n_clusters=k, init="k-means++")
-        kmeans.fit(data)
-        preds = kmeans.predict(data)    
-        silhouette = silhouette_score(data,preds)
-        km_silhouette.append(silhouette)
-    cluster_n = int(km_silhouette.index(max(km_silhouette)))+1
-    return cluster_n
+    def cluster(df, col):
+        data = pd.DataFrame()
+        for j in col:
+            data[j]=df[j]
+        cluster_n = silhoutte(data)
+        km = KMeans(n_clusters=cluster_n)
+        clusters = km.fit_predict(data)
+        df["label"] = clusters
+        fig = plt.figure(figsize=(20,10))
+        ax = fig.add_subplot(111, projection='3d')
+        colors = ['blue','red','yellow','orange','green','purple']
+        count = 0
+        for i in range(cluster_n):
+            ax.scatter(df[col[0]][df.label == i], df[col[1]][df.label == i], df[col[2]][df.label == i], c=colors[count], s=60)
+            count+=1
+        ax.view_init(30, 185)
+        plt.xlabel(col[0])
+        plt.ylabel(col[1])
+        ax.set_zlabel(col[2])
+        plt.show()
+
+    def silhoutte(data):
+        km_silhouette = []
+        for k in range(2,11):
+            kmeans = KMeans(n_clusters=k, init="k-means++")
+            kmeans.fit(data)
+            preds = kmeans.predict(data)    
+            silhouette = silhouette_score(data,preds)
+            km_silhouette.append(silhouette)
+        cluster_n = int(km_silhouette.index(max(km_silhouette)))+1
+        return cluster_n