changes done

Author: madhurprash

1_batch_transform_pipeline.ipynb

@@ -10,7 +10,7 @@
 import pandas as pd
 from ast import literal_eval
 from sklearn.ensemble import RandomForestClassifier
-
+ 
 from sklearn.metrics import (
     accuracy_score,
     precision_score,

code/evaluate.py

@@ -6,7 +6,7 @@
 import datetime
 import pandas as pd
 from io import StringIO
-
+ 
 ## define the logger
 logging.basicConfig(format='[%(asctime)s] p%(process)s {%(filename)s:%(lineno)d} %(levelname)s - %(message)s', level=logging.INFO)
 logger = logging.getLogger(__name__)

code/inference.py

@@ -6,7 +6,7 @@
 import pandas as pd
 from query import BATCH_INFERENCE_QUERY
 from presto_utils import fetch_data_from_presto
-
+ 
 ## define the logger
 logging.basicConfig(format='[%(asctime)s] p%(process)s {%(filename)s:%(lineno)d} %(levelname)s - %(message)s', level=logging.INFO)
 logger = logging.getLogger(__name__)

code/presto_preprocess_for_batch_inference.py

@@ -14,7 +14,7 @@
 import pandas as pd
 from query import TRAINING_DATA_QUERY
 from presto_utils import fetch_data_from_presto
-
+ 
 ## define the logger
 logging.basicConfig(format='[%(asctime)s] p%(process)s {%(filename)s:%(lineno)d} %(levelname)s - %(message)s', level=logging.INFO)
 logger = logging.getLogger(__name__)

code/presto_preprocess_for_training.py

@@ -11,7 +11,7 @@
 ## define the logger
 logging.basicConfig(format='[%(asctime)s] p%(process)s {%(filename)s:%(lineno)d} %(levelname)s - %(message)s', level=logging.INFO)
 logger = logging.getLogger(__name__)
-
+ 
 ## function to connect to the presto server
 def _connect_presto_server(args, username, password, catalog, schema):
     """

code/presto_utils.py

@@ -1,2 +1,2 @@
 ## Install the presto python client to connect to the database and extract query
-presto-python-client==0.8.4
+presto-python-client==0.8.4 

code/requirements.txt

@@ -6,7 +6,7 @@
 import logging
 import argparse
 import subprocess
-import numpy as np
+import numpy as np 
 import pandas as pd
 from typing import List
 from io import StringIO