comments

app.py

@@ -9,6 +9,8 @@ import numpy as np
 import hopsworks
 import joblib
 
+# Convert the input to the format the model expects
+
 
 def prepare_for_write(df):
     # Convert the categorical features to numerical
@@ -38,16 +40,17 @@ def prepare_for_write(df):
     return df
 
 
+# Login to hopsworks and get the feature store
 project = hopsworks.login()
 fs = project.get_feature_store()
 
-
+# Get the model from Hopsworks
 mr = project.get_model_registry()
 model = mr.get_model("titanic_modal", version=3)
 model_dir = model.download()
 model = joblib.load(model_dir + "/titanic_model.pkl")
 
-
+# For generating the input form
 catToInput = {
     "Sex": ["male", "female"],
     "Embarked": ["Southampton", "Cherbourg", "Queenstown"],
@@ -67,15 +70,9 @@ classToInput = {
 }
 
 
-# features = pd.read_csv(
-#     "https://raw.githubusercontent.com/Nathanotal/remoteFiles/main/titanicCleaned.csv")
-# features = features.drop(columns=["survived"])
-# featureLabels = features.columns
-featureLabels = ["Pclass", "Name", "Sex", "Age", "SibSp",
-                 "Parch", "Ticket", "Fare", "Cabin", "Embarked"]
 inputs = []
 numericalInputs = ["Age", "SibSp", "Parch", "Fare"]
-# Maybe move cabin to categorical
+# Maybe move cabin to categorical (or just remove it)
 worthlessInputs = ["Name", "Ticket", "Cabin", "Title"]
 categoricalInputs = ["Sex", "Embarked", "Pclass"]
 
@@ -84,6 +81,7 @@ columnHeaders = ["Pclass", "Sex", "Age", "SibSp",
 
 
 def titanic(Pclass, Sex, Age, SibSp, Parch, Fare, Embarked):
+    # Parse the unput and save it so we can run it through the model
     Embarked = cityToInput[Embarked]
     Pclass = classToInput[Pclass]
     # Create a dataframe from the input values
@@ -100,11 +98,16 @@ def titanic(Pclass, Sex, Age, SibSp, Parch, Fare, Embarked):
     # We add '[0]' to the result of the transformed 'res', because 'res' is a list, and we only want
     # the first element.
 
-    intLabelToText = {0: "Died", 1: "Survived"}
+    intLabelToText = {0: "Died", 1: "Survived"}  # Debug
 
     survived = res[0]
+    # The API we are using only supports this age range
+    if Age < 9:
+        Age = 9
+    if Age > 75:
+        Age = 75
 
-    # Todo: survivor, "https://fakeface.rest/face/json?maximum_age=50&gender=female&minimum_age=49"
+    # Generate a face of the inputted person
     generate_survivor_url = f'https://fakeface.rest/face/json?maximum_age={int(Age)}&gender={Sex}&minimum_age={int(Age)}'
     randomized_face_url = requests.get(
         generate_survivor_url).json()["image_url"]
@@ -112,7 +115,7 @@ def titanic(Pclass, Sex, Age, SibSp, Parch, Fare, Embarked):
     survivor_url = randomized_face_url
     img = Image.open(requests.get(survivor_url, stream=True).raw)
 
-    #
+    # Show a green check mark if the person is predicted to survive, otherwise show a red x
     red_cross_url = "https://www.iconsdb.com/icons/preview/red/x-mark-xxl.png"
     green_check_mark_url = "https://www.iconsdb.com/icons/preview/green/checkmark-xxl.png"
 
@@ -129,10 +132,11 @@ def titanic(Pclass, Sex, Age, SibSp, Parch, Fare, Embarked):
     return img, img2
 
 
+# All features present in the titanic dataset
 featureLabels = ["Pclass", "Name", "Sex", "Age", "SibSp",
                  "Parch", "Ticket", "Fare", "Cabin", "Embarked"]
 
-
+# Generate the input form
 for feature in featureLabels:
     if feature in numericalInputs:
         if feature == 'Age':
@@ -163,6 +167,7 @@ for feature in featureLabels:
     else:
         raise Exception(f'Feature: "{feature}" not found')
 
+# Create the interface
 demo = gr.Interface(
     fn=titanic,
     title="Titanic Survivor Predictive Analytics",