Update app.py

app.py

@@ -1,67 +1,48 @@
 import gradio as gr
+from PIL import Image
+import requests
 import hopsworks
+import joblib
 import pandas as pd
-import keras
 
 project = hopsworks.login()
 fs = project.get_feature_store()
 
+
 mr = project.get_model_registry()
-model = mr.get_model("wine_model", version=1)
+model = mr.get_model("iris_model", version=1)
 model_dir = model.download()
-
-n_input = 7 # Number of features
-n_hidden = 256  # Number of hidden nodes
-n_out = 3 # Number of classes
-model = keras.Sequential([
-    keras.layers.Dense(n_hidden, input_dim=n_input, activation='selu'),
-    keras.layers.BatchNormalization(),
-    keras.layers.Dense(n_hidden, activation='selu'),
-    keras.layers.BatchNormalization(),
-    keras.layers.Dense(n_hidden, activation='selu'),
-    keras.layers.BatchNormalization(),
-    keras.layers.Dense(n_hidden, activation='selu'),
-    keras.layers.BatchNormalization(),
-    keras.layers.Dense(n_hidden, activation='selu'),
-    keras.layers.Dense(n_out, activation='softmax')
-])
-
-model.load_weights(model_dir + '/wine_model.h5')
+model = joblib.load(model_dir + "/iris_model.pkl")
 print("Model downloaded")
 
-
-def wine(type, alcohol, density, sugar, vol_acid, chlorides, total_sulfur):
+def iris(sepal_length, sepal_width, petal_length, petal_width):
     print("Calling function")
-    type = 0 if type == 'White' else 1
-
 #     df = pd.DataFrame([[sepal_length],[sepal_width],[petal_length],[petal_width]], 
-    df = pd.DataFrame([[type, vol_acid, sugar, chlorides, total_sulfur, density, alcohol]],
-                      columns=['type', 'volatile acidity', 'sugar', 'chlorides', 'total sulfur dioxide', '´density', 'alcohol'])
+    df = pd.DataFrame([[sepal_length,sepal_width,petal_length,petal_width]], 
+                      columns=['sepal_length','sepal_width','petal_length','petal_width'])
     print("Predicting")
+    print(df)
     # 'res' is a list of predictions returned as the label.
-    wine_prediction = model.predict(df)
-    wine_prediction = {'High Quality': float(wine_prediction[0][2]),
-                       'Good Quality': float(wine_prediction[0][1]),
-                       'Low Quality': float(wine_prediction[0][0])}
-    return wine_prediction
-
-
+    res = model.predict(df) 
+    # We add '[0]' to the result of the transformed 'res', because 'res' is a list, and we only want 
+    # the first element.
+#     print("Res: {0}").format(res)
+    print(res)
+    flower_url = "https://raw.githubusercontent.com/featurestoreorg/serverless-ml-course/main/src/01-module/assets/" + res[0] + ".png"
+    img = Image.open(requests.get(flower_url, stream=True).raw)            
+    return img
+        
 demo = gr.Interface(
-    fn=wine,
-    title="Wine Quality Analytics",
-    description="Experiment with wine contents to predict what quality of wine it is.",
+    fn=iris,
+    title="Iris Flower Predictive Analytics",
+    description="Experiment with sepal/petal lengths/widths to predict which flower it is.",
     allow_flagging="never",
     inputs=[
-        gr.Radio(["White", "Red"], value='White', label="What kind of wine is it?"),
-        gr.Number(value=9.6, label="Alcohol content (%), normal range 8-15 %"),
-        gr.Number(value=0.9935, label="Density (kg/dm^3), normal range 0.99-0.101 kg/dm^3"),
-        gr.Number(value=1.1, label="Residual sugar content (g/L), normal range 0.6-66 g/L"),
-        gr.Number(value=0.26, label="Volatile acid content (g/L), normal range 0.1 - 1.6 g/L"),
-        gr.Number(value=0.04, label="Chloride content (g/L), normal range 0.01-0.6 g/L"),
-        gr.Number(value=147, label="Total sulfur dioxide content (ppm), normal range 6-450 ppm"),
+        gr.inputs.Number(default=2.0, label="sepal length (cm)"),
+        gr.inputs.Number(default=1.0, label="sepal width (cm)"),
+        gr.inputs.Number(default=2.0, label="petal length (cm)"),
+        gr.inputs.Number(default=1.0, label="petal width (cm)"),
         ],
-    outputs=gr.Label(num_top_classes=3))
+    outputs=gr.Image(type="pil"))
 
 demo.launch(debug=True)
-
-