added more ap idata validation

ml/src/api/main.py

@@ -169,12 +169,20 @@ def predict(incoming_data: InputData):
     ats_resolution = settings['ats_resolution']
     
     data = incoming_data.dict()
-    
-    incoming_ats = [x.strip(' ') for x in data['Ats'].split(",")]
-    ats_converter = {'ats_id': str}
-    ats_df = read_dataframe('ats.csv', converters=ats_converter)
-    if all(x in list(ats_df['ats_id']) for x in incoming_ats) != True:
-        raise HTTPException(status_code=400, detail="Ats not found, check ats list")
+    if not data['Gender'] in [0, 1]:
+        raise HTTPException(status_code=400, detail="Invalid gender, check input")
+    if not data['BirthYear'] in range(100-99, 100+99+1):
+        raise HTTPException(status_code=400, detail="Invalid birth year, check input")
+    if not data['LoanPeriod'] >= 0:
+        raise HTTPException(status_code=400, detail="Invalid loan period, check input")
+    if data['Ats'] != '':
+        incoming_ats = [x.strip(' ') for x in data['Ats'].split(",")]
+        ats_converter = {'ats_id': str}
+        ats_df = read_dataframe('ats.csv', converters=ats_converter)
+        if all(x in list(ats_df['ats_id']) for x in incoming_ats) != True:
+            raise HTTPException(status_code=400, detail="Ats not found, check ats list")
+    else:
+        data['Ats'] = '0'
     
     df = prepare_data(data, ats_resolution)
     arguments = generate_arguments(df, ats_resolution)
@@ -242,7 +250,7 @@ def generate_arguments(df: pd.DataFrame, ats_resolution: int):
 
 def load_settings(file_name):
     dir_path = os.path.dirname(os.path.realpath(__file__))
-    with open(Path.joinpath(dir_path, file_name), 'r') as stream:
+    with open(f'{dir_path}/{file_name}', 'r') as stream:
         settings = yaml.safe_load(stream)
     return settings
 
@@ -255,7 +263,12 @@ def get_ats_name_from_hmi(ats_id: str):
 def prepare_data(data: dict, ats_resolution: int) -> pd.DataFrame:
     new_data = {k: [v] for k, v in data.items()}
     new_data_df = pd.DataFrame.from_dict(new_data)
-    new_data_df['NumberAts'] = len(new_data_df['Ats'][0].split(","))
+    
+    if new_data_df['Ats'][0] == '0':
+        new_data_df['NumberAts'] = 0
+    else:
+        ats = list(filter(None, new_data_df['Ats'][0].split(",")))
+        new_data_df['NumberAts'] = len(ats)
     
     df = split_categorical_columns(new_data_df, col='Ats', tag='Ats',
                                    resolution=ats_resolution)