Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -9,13 +9,6 @@ import matplotlib.pyplot as plt
|
|
9 |
from io import BytesIO
|
10 |
|
11 |
|
12 |
-
|
13 |
-
# num_imputer = joblib.load('C:/Users/user/Desktop/Churn_Prediction_ML/models/numerical_imputer.joblib')
|
14 |
-
# cat_imputer = joblib.load('C:/Users/user/Desktop/Churn_Prediction_ML/models/cat_imputer.joblib')
|
15 |
-
# encoder = joblib.load('C:/Users/user/Desktop/Churn_Prediction_ML/models/encoder.joblib')
|
16 |
-
# scaler = joblib.load('C:/Users/user/Desktop/Churn_Prediction_ML/models/scaler.joblib')
|
17 |
-
# lr_model = joblib.load('C:/Users/user/Desktop/Churn_Prediction_ML/models/lr_smote_model.joblib')
|
18 |
-
|
19 |
num_imputer = joblib.load('numerical_imputer.joblib')
|
20 |
cat_imputer = joblib.load('cat_imputer.joblib')
|
21 |
encoder = joblib.load('encoder.joblib')
|
@@ -57,7 +50,7 @@ original_feature_names = ['MONTANT', 'FREQUENCE_RECH', 'REVENUE', 'ARPU_SEGMENT'
|
|
57 |
st.set_page_config(layout="wide")
|
58 |
|
59 |
# Main page - Churn Prediction
|
60 |
-
st.title('CUSTOMER CHURN PREDICTION APP
|
61 |
|
62 |
# Main page - Churn Prediction
|
63 |
st.markdown("Churn is a one of the biggest problem in the telecom industry. Research has shown that the average monthly churn rate among the top 4 wireless carriers in the US is 1.9% - 2%")
|
@@ -82,10 +75,10 @@ model_type = models[model_name]['type']
|
|
82 |
|
83 |
|
84 |
# Collect input from the user
|
85 |
-
st.sidebar.title('
|
86 |
input_features = {
|
87 |
'MONTANT': st.sidebar.number_input('Top-up Amount (MONTANT)'),
|
88 |
-
'FREQUENCE_RECH': st.sidebar.number_input('
|
89 |
'REVENUE': st.sidebar.number_input('Monthly income of the client (REVENUE)'),
|
90 |
'ARPU_SEGMENT': st.sidebar.number_input('Income over 90 days / 3 (ARPU_SEGMENT)'),
|
91 |
'FREQUENCE': st.sidebar.number_input('Number of times the client has made an income (FREQUENCE)'),
|
@@ -208,7 +201,6 @@ if st.sidebar.button('Predict Churn'):
|
|
208 |
"- It\s crucial to genuinely address their concerns and provide value to rebuild their trust in your business\n"
|
209 |
"- Regularly evaluate the effectiveness of your strategies and adjust them as needed based on customer responses and feedback\n")
|
210 |
else:
|
211 |
-
#churn_index = churn_indices["No Churn"]
|
212 |
churn_prob = churn_probability[churn_index]
|
213 |
with col1:
|
214 |
st.success(f"This customer is not likely to churn with a probability of {churn_prob * 100:.2f}% 😀")
|
@@ -296,7 +288,7 @@ if st.sidebar.button('Predict Churn'):
|
|
296 |
else:
|
297 |
churn_comparison = "equal"
|
298 |
|
299 |
-
explanation = f"This
|
300 |
f"with the average churn rate of all customers. It provides insights into how the " \
|
301 |
f"individual customer's churn likelihood ({predicted_churn_prob:.2f}) compares to the " \
|
302 |
f"overall trend. The 'Churn Probability' represents the likelihood of churn " \
|
|
|
9 |
from io import BytesIO
|
10 |
|
11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
num_imputer = joblib.load('numerical_imputer.joblib')
|
13 |
cat_imputer = joblib.load('cat_imputer.joblib')
|
14 |
encoder = joblib.load('encoder.joblib')
|
|
|
50 |
st.set_page_config(layout="wide")
|
51 |
|
52 |
# Main page - Churn Prediction
|
53 |
+
st.title('CUSTOMER CHURN PREDICTION APP')
|
54 |
|
55 |
# Main page - Churn Prediction
|
56 |
st.markdown("Churn is a one of the biggest problem in the telecom industry. Research has shown that the average monthly churn rate among the top 4 wireless carriers in the US is 1.9% - 2%")
|
|
|
75 |
|
76 |
|
77 |
# Collect input from the user
|
78 |
+
st.sidebar.title('ENTER CUSTOMER DETAILS')
|
79 |
input_features = {
|
80 |
'MONTANT': st.sidebar.number_input('Top-up Amount (MONTANT)'),
|
81 |
+
'FREQUENCE_RECH': st.sidebar.number_input('No. of Times the Customer Refilled (FREQUENCE_RECH)'),
|
82 |
'REVENUE': st.sidebar.number_input('Monthly income of the client (REVENUE)'),
|
83 |
'ARPU_SEGMENT': st.sidebar.number_input('Income over 90 days / 3 (ARPU_SEGMENT)'),
|
84 |
'FREQUENCE': st.sidebar.number_input('Number of times the client has made an income (FREQUENCE)'),
|
|
|
201 |
"- It\s crucial to genuinely address their concerns and provide value to rebuild their trust in your business\n"
|
202 |
"- Regularly evaluate the effectiveness of your strategies and adjust them as needed based on customer responses and feedback\n")
|
203 |
else:
|
|
|
204 |
churn_prob = churn_probability[churn_index]
|
205 |
with col1:
|
206 |
st.success(f"This customer is not likely to churn with a probability of {churn_prob * 100:.2f}% 😀")
|
|
|
288 |
else:
|
289 |
churn_comparison = "equal"
|
290 |
|
291 |
+
explanation = f"This compares the churn probability of the selected customer " \
|
292 |
f"with the average churn rate of all customers. It provides insights into how the " \
|
293 |
f"individual customer's churn likelihood ({predicted_churn_prob:.2f}) compares to the " \
|
294 |
f"overall trend. The 'Churn Probability' represents the likelihood of churn " \
|