app fix

app.py

@@ -1,36 +1,116 @@
 import gradio as gr
-import torch
-from torch import tensor
-from torch.nn import functional as F
-import pandas as pd
+import numpy as np
+from joblib import load
 
-# model = torch.load('model.pkl')
+rf = load("model.pkl")
 
-def calc(X_test):
-    # return model.predict(X_test)
-    return X_test
+columns = [
+    "BMI",
+    "Smoking",
+    "AlcoholDrinking",
+    "Stroke",
+    "PhysicalHealth",
+    "MentalHealth",
+    "DiffWalking",
+    "Sex",
+    "AgeEstimate",
+    "Race",
+    "Diabetic",
+    "PhysicalActivity",
+    "GenHealth",
+    "SleepTime",
+    "Asthma",
+    "KidneyDisease",
+    "SkinCancer"
+]
 
-# def main(inputs):
-#     df = pd.DataFrame(columns=inputs[1:])
-#     df.loc[0] = inputs[1:]
 
-#     for column in df.columns:
-#         df[column] = label_encoder.fit_transform(df[column])
+def predict(
+    BMI,
+    Smoking,
+    AlcoholDrinking,
+    Stroke,
+    PhysicalHealth,
+    MentalHealth,
+    DiffWalking,
+    Sex,
+    AgeEstimate,
+    Race,
+    Diabetic,
+    PhysicalActivity,
+    GenHealth,
+    SleepTime,
+    Asthma,
+    KidneyDisease,
+    SkinCancer
+):
+    data = np.array(
+        [
+            [
+                BMI,
+                Smoking,
+                AlcoholDrinking,
+                Stroke,
+                PhysicalHealth,
+                MentalHealth,
+                DiffWalking,
+                Sex,
+                AgeEstimate,
+                Race+1,
+                Diabetic,
+                PhysicalActivity,
+                GenHealth+1,
+                SleepTime,
+                Asthma,
+                KidneyDisease,
+                SkinCancer
+            ]
+        ]
+    )
+    pred = rf.predict(data)[0]
+    return {"HeartDisease": pred}
 
-#     t_indep = tensor(df[indep_cols].values, dtype=torch.float)
-#     vals,indices = t_indep.max(dim=0)
-#     t_indep = t_indep / vals
-#     return calc_preds(coeffs, t_indep)
+
+inputs = [
+    gr.Slider(minimum=0, maximum=150, label="BMI"),
+    gr.Checkbox(label="Smoking"),
+    gr.Checkbox(label="Alcohol Drinking"),
+    gr.Checkbox(label="Stroke"),
+    gr.Number(label="Physical Health"),
+    gr.Number(label="MentalHealth"),
+    gr.Checkbox(label="Diff Walking"),
+    gr.Dropdown(["Female", "Male"], type="index", label="Sex"),
+    gr.Dropdown(['18-24',
+            '25-29',
+            '30-34',
+            '35-39',
+            '40-44',
+            '45-49',
+            '50-54',
+            '55-59',
+            '60-64',
+            '65-69',
+            '70-74',
+            '75-79',
+            '80 or older'], type="index", label="Age Category"),
+    gr.Dropdown(["White", "Black", "Asian", "American Indian/Alaskan Native", "Hispanic", "Other"], type="index", label="Race"),
+    gr.Dropdown(["No", "Yes", "Yes(during pregnancy)", "No, borderline diabetes"], type="index", label="Diebetic"),
+    gr.Checkbox(label="Physical Activity"),
+    gr.Dropdown(["Poor", "Fair", "Good", "Very good", "Excellent"], type="index", label="General Health"),
+    gr.Number(label="Sleep time"),
+    gr.Checkbox(label="Asthma"),
+    gr.Checkbox(label="Kidney Disease"),
+    gr.Checkbox(label="Skin Cancer"),
+]
+
+
+output = gr.Label(num_top_classes=1)
 
 iface = gr.Interface(
-    fn=calc,
-    inputs=[gr.Dropdown('Feature 1', choices=['Option 1', 'Option 2', 'Option 3']),
-            gr.Dropdown('Feature 2', choices=['Option A', 'Option B', 'Option C']),
-            # Add more DropDown or other input components as needed
-            ],
-    outputs="text",
-    title="Heart Disease identifier",
-    description="Identifies if a person has/will have a heart disease"
+    fn=predict,
+    inputs=inputs,
+    outputs=output,
+    description="The purpose of this model is to predict whether or not a person has any heart diseases or not.",
 )
 
-iface.launch()
+iface.launch()