lab5done[JOAO ALMEIDA]

LAB5done.py

@@ -0,0 +1,219 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# In[2]:
+
+
+import pandas as pd
+
+
+# In[3]:
+
+
+import numpy as np
+
+
+# In[4]:
+
+
+df = pd.read_csv(r"C:\Users\joaof\ironhack\labs\lab-customer-analysis-round-2\files_for_lab\csv_files\marketing_customer_analysis.csv")
+
+
+# In[5]:
+
+
+df = pd.read_csv(r"C:\Users\joaof\ironhack\labs\lab-customer-analysis-round-4\files_for_lab\csv_files\marketing_customer_analysis.csv")
+
+
+# In[6]:
+
+
+df
+
+
+# In[7]:
+
+
+df.shape
+
+
+# In[8]:
+
+
+df.head(5)
+
+
+# In[9]:
+
+
+cols = []
+
+
+# In[10]:
+
+
+for i in range(len(df.columns)):
+    cols.append(df.columns[i].lower().replace(' ', '_'))
+
+
+# In[11]:
+
+
+df.columns = cols
+
+
+# In[12]:
+
+
+df
+
+
+# In[13]:
+
+
+numeric_columns = df.select_dtypes(include=['int', 'float'])
+
+
+# In[14]:
+
+
+numeric_columns
+
+
+# In[15]:
+
+
+categorical_columns = df.select_dtypes(include= ['object'])
+
+
+# In[16]:
+
+
+categorical_columns
+
+
+# In[17]:
+
+
+import seaborn as sns
+
+
+# In[18]:
+
+
+import matplotlib.pyplot as plt
+
+
+# In[19]:
+
+
+sns.histplot(data=numeric_columns, x = 'income')
+plt.ylabel('frequency')
+plt.xlabel('income')
+plt.show()
+
+
+# In[20]:
+
+
+sns.histplot(data=numeric_columns, x = 'monthly_premium_auto')
+plt.ylabel('frequency')
+plt.xlabel('monltlhy_premium_auto')
+plt.show()
+
+
+# In[21]:
+
+
+sns.histplot(data=numeric_columns, x = 'income')
+plt.ylabel('frequency')
+plt.xlabel('monltlhy_premium_auto')
+plt.show()
+
+
+# In[22]:
+
+
+sns.displot(data=numeric_columns, x = 'months_since_last_claim')
+plt.ylabel('frequency')
+plt.xlabel('Months Since Last Claim')
+plt.show()
+
+
+# In[23]:
+
+
+sns.displot(data=numeric_columns, x = 'number_of_policies')
+plt.ylabel('frequency')
+plt.xlabel('Number of open complaints')
+plt.show()
+
+
+# In[24]:
+
+
+sns.displot(data=numeric_columns, x = 'number_of_open_complaints')
+plt.ylabel('frequency')
+plt.xlabel('Number of open complaints')
+plt.show()
+
+
+# In[25]:
+
+
+correlation_matrix =  numeric_columns.corr()
+
+
+# In[26]:
+
+
+correlation_matrix
+
+
+# In[27]:
+
+
+sns.heatmap(correlation_matrix, annot=True)
+plt.show()
+mask = np.zeros_like(correlation_matrix)
+mask[np.triu_indices_from(mask)] = True 
+fig, ax = plt.subplots(figsize=(10, 8))
+ax = sns.heatmap(correlation_matrix, mask=mask, annot=True)
+plt.show()
+
+
+# In[28]:
+
+
+from sklearn.model_selection import train_test_split
+
+
+# In[29]:
+
+
+df_train, df_test = train_test_split(df, test_size=0.2, random_state=100)
+print(df_train.shape, df_test.shape)
+
+
+# In[31]:
+
+
+#X-Y SPLIT
+X = df.drop('customer_lifetime_value', axis=1)
+Y= df['customer_lifetime_value']
+
+
+# In[32]:
+
+
+print("Features (X):")
+print(X)
+print("\nTarget (Y):")
+print(Y)
+
+
+# In[ ]:
+
+
+
+