Resolved numpy, matplotlib version incompatibiltiy issues for report generation

Author: VishalKumar-S

CML_Reports/predictions_scatter_plot.png

@@ -1,12 +1,12 @@
 scikit-learn
-numpy<2
+numpy<2.0.0
 pandas
 zenml
 h2o
 evidently
 streamlit
 neptune
-matplotlib==3.4.3
+matplotlib
 email_validator
 discord.py
 slack_sdk

Evidently_Reports/data_quality_suite.html

@@ -1,4 +1,5 @@
 import pandas as pd
+import numpy as np 
 
 class DataPreprocessor:
     def __init__(self, data):
@@ -20,7 +21,7 @@ def drop_columns(self, data, columns):
         return data.drop(columns, axis=1)
 
     def calculate_cpc(self, data):
-        data['CPC'] = data['Spent'] / data['Clicks']
+        data['CPC'] = np.where(data['Clicks']>0, data['Spent'] / data['Clicks'],0)
         return data['CPC']
 
     def impute_missing_values(self, data, columns):

Production_data/user_uploaded_dataset.csv

@@ -45,7 +45,7 @@ def data_quality_validation(curr_data: pd.DataFrame, user_email: str) -> pd.Data
 
     threshold = passed_tests / total_tests if total_tests > 0 else 0
 
-    if threshold < 0.85:
+    if threshold > 0.85:
 
         logging.error("Data quality tests got failed. Logging failed reports and sending alerts...")
         # Initialize a run