initial

.gitignore

@@ -0,0 +1 @@
+.idea/*

data/model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de759f7c4ad50a7d58a7531056fcda31314b13a9948a613b167fa35969290c14
+size 164923921

main.py

@@ -0,0 +1,88 @@
+import gradio as gr
+import pandas as pd
+import joblib
+
+import matplotlib
+
+matplotlib.use('TkAgg')
+
+
+def predict(gender, bmi, asthma, glucose, pulse):
+
+    gender_char = 1
+    if gender == "Female":
+        gender_char = 0
+
+    asthma_val = False
+    if asthma is True:
+        asthma_val = 1
+
+    pred_data = {
+        "id": 0,
+        "rcount": train_data['rcount'].mean(),
+        "gender": gender_char,
+        "dialysisrenalendstage": False,
+        "asthma": asthma_val,
+        "irondef": False,
+        "pneum": False,
+        "substancedependence": False,
+        "psychologicaldisordermajor": False,
+        "depress": False,
+        "psychother": False,
+        "fibrosisandother": False,
+        "malnutrition": False,
+        "hemo": False,
+        "hematocrit": False,
+        "neutrophils": False,
+        "sodium": train_data['sodium'].mean(),
+        "glucose": glucose,
+        "bloodureanitro": train_data['bloodureanitro'].mean(),
+        "creatinine": train_data['creatinine'].mean(),
+        "bmi": bmi,
+        "pulse": pulse,
+        "respiration": train_data['respiration'].mean(),
+        "secondarydiagnosisnonicd9": 0,
+        "facid": 0
+    }
+
+    features = ['id', 'rcount', 'gender', 'dialysisrenalendstage', 'asthma', 'irondef', 'pneum', 'substancedependence',
+                'psychologicaldisordermajor', 'depress', 'psychother', 'fibrosisandother', 'malnutrition', 'hemo',
+                'hematocrit', 'neutrophils', 'sodium', 'glucose', 'bloodureanitro', 'creatinine', 'bmi', 'pulse',
+                'respiration', 'secondarydiagnosisnonicd9', 'facid']
+
+    pred_df = pd.DataFrame([pred_data], columns=features)
+    result = "Error in prediction"
+
+    try:
+        model = joblib.load('data/model.pkl')
+        result = model.predict(pred_df)
+    except Exception as e:
+        print(f"Fehler beim Laden des Modells: {str(e)}")
+
+    print(result)
+    return result
+
+
+gender = 0
+bmi_min, bmi_max, bmi_default = 15, 50, 25
+asthma = 0
+glucose_min, glucose_max, glucose_default = 30, 190, 85
+pulse_min, pulse_max, pulse_default = 40, 70, 180
+
+# Create the interface
+iface = gr.Interface(
+    fn=predict,
+    inputs=[
+        gr.components.Radio(["Female", "Male"], label="Gender"),
+        gr.components.Slider(minimum=bmi_min, maximum=bmi_max, value=bmi_default, label="BMI"),
+        gr.components.Checkbox(label="Asthma"),
+        gr.components.Slider(minimum=glucose_min, maximum=glucose_max, value=glucose_default, label="Glucose Level"),
+        gr.components.Slider(minimum=pulse_min, maximum=glucose_max, value=pulse_default, label="Pulse")
+    ],
+    outputs=gr.components.Textbox(label="Prediction of stay length"),
+    title="Length of Stay - Predictor",
+    description="""Enter the values to get a prediction of your length of stay""",
+)
+
+# Launch the interface
+iface.launch()