Upload app.py

app.py

@@ -235,20 +235,24 @@ def y1_predict_rf(*args):
 def y2_predict_xgb(*args):
     df2 = pd.DataFrame([args], columns=x2.columns)
     df2 = df2.astype({col: "category" for col in categorical_columns2})
-    pos_pred = y2_model_xgb.predict(xgb.DMatrix(df2, enable_categorical=True))
-    return {"Facility Discharge": float(pos_pred[0]), "Home Discharge": 1 - float(pos_pred[0])}
+    d2 = dict.fromkeys(df2.select_dtypes(np.int64).columns, np.int32)
+    df2 = df2.astype(d2)
+    pos_pred = y2_model_xgb.predict_proba(df2)
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 def y2_predict_lgb(*args):
-    df2 = pd.DataFrame([args], columns=x2_lgb.columns)
+    df2 = pd.DataFrame([args], columns=x2.columns)
     df2 = df2.astype({col: "category" for col in categorical_columns2})
-    pos_pred = y2_model_lgb.predict(df2)
-    return {"Facility Discharge": float(pos_pred[0]), "Home Discharge": 1 - float(pos_pred[0])}
+    d2 = dict.fromkeys(df2.select_dtypes(np.int64).columns, np.int32)
+    df2 = df2.astype(d2)
+    pos_pred = y2_model_lgb.predict_proba(df2)
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 def y2_predict_cb(*args):
     df2 = pd.DataFrame([args], columns=x2.columns)
     df2 = df2.astype({col: "category" for col in categorical_columns2})
     pos_pred = y2_model_cb.predict(Pool(df2, cat_features = categorical_columns2), prediction_type='Probability')
-    return {"Facility Discharge": float(pos_pred[0][1]), "Home Discharge": float(pos_pred[0][0])}
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 def y2_predict_rf(*args):
     df2 = pd.DataFrame([args], columns=x2_rf.columns)
@@ -256,89 +260,100 @@ def y2_predict_rf(*args):
     d2 = dict.fromkeys(df2.select_dtypes(np.int64).columns, np.int32)
     df2 = df2.astype(d2)
     pos_pred = y2_model_rf.predict_proba(df2)
-    return {"Facility Discharge": float(pos_pred[0][1]), "Home Discharge": float(pos_pred[0][0])}
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 #Define predict for y3 (LOS).
 def y3_predict_xgb(*args):
     df3 = pd.DataFrame([args], columns=x3.columns)
     df3 = df3.astype({col: "category" for col in categorical_columns3})
-    pos_pred = y3_model_xgb.predict(xgb.DMatrix(df3, enable_categorical=True))
-    return {"Prolonged LOS": float(pos_pred[0]), "No Prolonged LOS": 1 - float(pos_pred[0])}
+    d3 = dict.fromkeys(df3.select_dtypes(np.int64).columns, np.int33)
+    df3 = df3.astype(d3)
+    pos_pred = y3_model_xgb.predict_proba(df3)
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 def y3_predict_lgb(*args):
-    df3 = pd.DataFrame([args], columns=x3_lgb.columns)
+    df3 = pd.DataFrame([args], columns=x3.columns)
     df3 = df3.astype({col: "category" for col in categorical_columns3})
-    pos_pred = y3_model_lgb.predict(df3)
-    return {"Prolonged LOS": float(pos_pred[0]), "No Prolonged LOS": 1 - float(pos_pred[0])}
+    d3 = dict.fromkeys(df3.select_dtypes(np.int64).columns, np.int33)
+    df3 = df3.astype(d3)
+    pos_pred = y3_model_lgb.predict_proba(df3)
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 def y3_predict_cb(*args):
     df3 = pd.DataFrame([args], columns=x3.columns)
     df3 = df3.astype({col: "category" for col in categorical_columns3})
     pos_pred = y3_model_cb.predict(Pool(df3, cat_features = categorical_columns3), prediction_type='Probability')
-    return {"Prolonged LOS": float(pos_pred[0][1]), "No Prolonged LOS": float(pos_pred[0][0])}
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 def y3_predict_rf(*args):
     df3 = pd.DataFrame([args], columns=x3_rf.columns)
     df3 = df3.astype({col: "category" for col in categorical_columns3})
-    d3 = dict.fromkeys(df3.select_dtypes(np.int64).columns, np.int32)
-    df3 = df.astype(d3)
+    d3 = dict.fromkeys(df3.select_dtypes(np.int64).columns, np.int33)
+    df3 = df3.astype(d3)
     pos_pred = y3_model_rf.predict_proba(df3)
-    return {"Prolonged LOS": float(pos_pred[0][1]), "No Prolonged LOS": float(pos_pred[0][0])}
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 #Define predict for y4 (ICU LOS).
 def y4_predict_xgb(*args):
     df4 = pd.DataFrame([args], columns=x4.columns)
     df4 = df4.astype({col: "category" for col in categorical_columns4})
-    pos_pred = y4_model_xgb.predict(xgb.DMatrix(df4, enable_categorical=True))
-    return {"Prolonged ICU LOS": float(pos_pred[0]), "No Prolonged ICU LOS": 1 - float(pos_pred[0])}
+    d4 = dict.fromkeys(df4.select_dtypes(np.int64).columns, np.int34)
+    df4 = df4.astype(d4)
+    pos_pred = y4_model_xgb.predict_proba(df4)
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 def y4_predict_lgb(*args):
-    df4 = pd.DataFrame([args], columns=x4_lgb.columns)
+    df4 = pd.DataFrame([args], columns=x4.columns)
     df4 = df4.astype({col: "category" for col in categorical_columns4})
-    pos_pred = y4_model_lgb.predict(df4)
-    return {"Prolonged ICU LOS": float(pos_pred[0]), "No Prolonged ICU LOS": 1 - float(pos_pred[0])}
+    d4 = dict.fromkeys(df4.select_dtypes(np.int64).columns, np.int34)
+    df4 = df4.astype(d4)
+    pos_pred = y4_model_lgb.predict_proba(df4)
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 def y4_predict_cb(*args):
-    df4 = df4.DataFrame([args], columns=x4.columns)
+    df4 = pd.DataFrame([args], columns=x4.columns)
     df4 = df4.astype({col: "category" for col in categorical_columns4})
     pos_pred = y4_model_cb.predict(Pool(df4, cat_features = categorical_columns4), prediction_type='Probability')
-    return {"Prolonged ICU LOS": float(pos_pred[0][1]), "No Prolonged ICU LOS": float(pos_pred[0][0])}
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 def y4_predict_rf(*args):
     df4 = pd.DataFrame([args], columns=x4_rf.columns)
     df4 = df4.astype({col: "category" for col in categorical_columns4})
-    d4 = dict.fromkeys(df4.select_dtypes(np.int64).columns, np.int32)
+    d4 = dict.fromkeys(df4.select_dtypes(np.int64).columns, np.int34)
     df4 = df4.astype(d4)
     pos_pred = y4_model_rf.predict_proba(df4)
-    return {"Prolonged ICU LOS": float(pos_pred[0][1]), "No Prolonged ICU LOS": float(pos_pred[0][0])}
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 #Define predict for y5 (complications).
-def y5_predict_xgb(*args):
-    df5 = pd.DataFrame([args], columns=x5.columns)
-    df5 = df5.astype({col: "category" for col in categorical_columns5})
-    pos_pred = y5_model_xgb.predict(xgb.DMatrix(df5, enable_categorical=True))
-    return {"Major Complications": float(pos_pred[0]), "No Major Complications": 1 - float(pos_pred[0])}
-
-def y5_predict_lgb(*args):
-    df5 = pd.DataFrame([args], columns=x5_lgb.columns)
-    df5 = df5.astype({col: "category" for col in categorical_columns5})
-    pos_pred = y5_model_lgb.predict(df5)
-    return {"Major Complications": float(pos_pred[0]), "No Major Complications": 1 - float(pos_pred[0])}
+def y4_predict_xgb(*args):
+    df4 = pd.DataFrame([args], columns=x4.columns)
+    df4 = df4.astype({col: "category" for col in categorical_columns4})
+    d4 = dict.fromkeys(df4.select_dtypes(np.int64).columns, np.int34)
+    df4 = df4.astype(d4)
+    pos_pred = y4_model_xgb.predict_proba(df4)
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
-def y5_predict_cb(*args):
-    df5 = pd.DataFrame([args], columns=x5.columns)
-    df5 = df5.astype({col: "category" for col in categorical_columns5})
-    pos_pred = y5_model_cb.predict(Pool(df5, cat_features = categorical_columns5), prediction_type='Probability')
-    return {"Major Complications": float(pos_pred[0][1]), "No Major Complications": float(pos_pred[0][0])}
+def y4_predict_lgb(*args):
+    df4 = pd.DataFrame([args], columns=x4.columns)
+    df4 = df4.astype({col: "category" for col in categorical_columns4})
+    d4 = dict.fromkeys(df4.select_dtypes(np.int64).columns, np.int34)
+    df4 = df4.astype(d4)
+    pos_pred = y4_model_lgb.predict_proba(df4)
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
-def y5_predict_rf(*args):
-    df5 = pd.DataFrame([args], columns=x5_rf.columns)
-    df5 = df.astype({col: "category" for col in categorical_columns5})
-    d5 = dict.fromkeys(df5.select_dtypes(np.int64).columns, np.int32)
-    df5 = df5.astype(d5)
-    pos_pred = y5_model_rf.predict_proba(df5)
-    return {"Major Complications": float(pos_pred[0][1]), "No Major Complications": float(pos_pred[0][0])}
+def y4_predict_cb(*args):
+    df4 = pd.DataFrame([args], columns=x4.columns)
+    df4 = df4.astype({col: "category" for col in categorical_columns4})
+    pos_pred = y4_model_cb.predict(Pool(df4, cat_features = categorical_columns4), prediction_type='Probability')
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
+def y4_predict_rf(*args):
+    df4 = pd.DataFrame([args], columns=x4_rf.columns)
+    df4 = df4.astype({col: "category" for col in categorical_columns4})
+    d4 = dict.fromkeys(df4.select_dtypes(np.int64).columns, np.int34)
+    df4 = df4.astype(d4)
+    pos_pred = y4_model_rf.predict_proba(df4)
+    return {"Mortality": float(pos_pred[0][1]), "No Mortality": float(pos_pred[0][0])}
 
 #Define function for wrapping feature labels.
 def wrap_labels(ax, width, break_long_words=False):
@@ -406,7 +421,7 @@ def y1_interpret_cb(*args):
     return fig
 
 def y1_interpret_rf(*args):
-    df1 = pd.DataFrame([args], columns=x1_rf.columns)
+    df1 = pd.DataFrame([args], columns=x1.columns)
     df1 = df1.astype({col: "category" for col in categorical_columns1})
     shap_values1 = y1_explainer_rf.shap_values(df1)
     shap_values1 = np.abs(shap_values1)
@@ -479,7 +494,7 @@ def y2_interpret_cb(*args):
     return fig
 
 def y2_interpret_rf(*args):
-    df2 = pd.DataFrame([args], columns=x2_rf.columns)
+    df2 = pd.DataFrame([args], columns=x2.columns)
     df2 = df2.astype({col: "category" for col in categorical_columns2})
     shap_values2 = y2_explainer_rf.shap_values(df2)
     shap_values2 = np.abs(shap_values2)
@@ -552,7 +567,7 @@ def y3_interpret_cb(*args):
     return fig
 
 def y3_interpret_rf(*args):
-    df3 = pd.DataFrame([args], columns=x3_rf.columns)
+    df3 = pd.DataFrame([args], columns=x3.columns)
     df3 = df3.astype({col: "category" for col in categorical_columns3})
     shap_values3 = y3_explainer_rf.shap_values(df3)
     shap_values3 = np.abs(shap_values3)
@@ -627,7 +642,7 @@ def y4_interpret_cb(*args):
 def y4_interpret_rf(*args):
     df4 = pd.DataFrame([args], columns=x4_rf.columns)
     df4 = df4.astype({col: "category" for col in categorical_columns4})
-    shap_values4 = y4_explainer_rf.shap_values(df4)
+    shap_values4 = y4_explainer.shap_values(df4)
     shap_values4 = np.abs(shap_values4)
     shap.bar_plot(shap_values4[0][0], max_display = 10, show = False, feature_names = f4_names)
     fig = plt.gcf()
@@ -698,7 +713,7 @@ def y5_interpret_cb(*args):
     return fig
 
 def y5_interpret_rf(*args):
-    df5 = pd.DataFrame([args], columns=x5_rf.columns)
+    df5 = pd.DataFrame([args], columns=x5.columns)
     df5 = df5.astype({col: "category" for col in categorical_columns5})
     shap_values = y5_explainer_rf.shap_values(df5)
     shap_values1 = np.abs(shap_values5)