Update with new results

app.py

@@ -6,7 +6,7 @@ from utils.prediction import predict_drug
 
 # Loading the model
 filename = "./model/drug_pipeline.sav"
-pipe = pickle.load(open(filename, 'rb'))
+pipe = pickle.load(open(filename, "rb"))
 
 
 # Web interface section
@@ -27,15 +27,22 @@ with st.sidebar.expander("Single Prediction"):
         cholesterol_list = np.array(["HIGH", "NORMAL"])
         cholesterol = st.radio("Select your Cholesterol", cholesterol_list)
 
-        Na_to_k = st.number_input(label="NA_to_K", min_value=6.2, step=0.1, max_value=38.2)
+        Na_to_k = st.number_input(
+            label="NA_to_K", min_value=6.2, step=0.1, max_value=38.2
+        )
         submit_button = st.form_submit_button(label="Predict")
 
 if submit_button:
-    prediction = predict_drug(age, gender[0], blood_pressure, cholesterol, Na_to_k, pipe)
+    prediction = predict_drug(
+        age, gender[0], blood_pressure, cholesterol, Na_to_k, pipe
+    )
 
     if prediction is None:
         st.error("An error occurred while getting the prediction!")
 
     message = f"The Drug is {prediction}!"
-    message_color = 'red' if prediction == 1 else 'green'
-    st.markdown(f"<h3 style='text-align: left;color:{message_color}'> {(message)} </h3>", unsafe_allow_html=True)
+    message_color = "red" if prediction == 1 else "green"
+    st.markdown(
+        f"<h3 style='text-align: left;color:{message_color}'> {(message)} </h3>",
+        unsafe_allow_html=True,
+    )

results/metrics.txt

@@ -1,2 +1,2 @@
 
-Accuracy=0.95, F1_score = 0.89
+Accuracy=0.97, F1_score = 0.85

train.py

@@ -17,7 +17,9 @@ from sklearn.model_selection import train_test_split
 X = drug_df.drop("Drug", axis=1).values
 y = drug_df.Drug.values
 
-X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=125)
+X_train, X_test, y_train, y_test = train_test_split(
+    X, y, test_size=0.3, random_state=125
+)
 
 # Pipeline
 cat_col = [1, 2, 3]
@@ -32,7 +34,7 @@ transform = ColumnTransformer(
 )
 
 pipe = Pipeline(
-    steps = [
+    steps=[
         ("preprocessing", transform),
         ("model", RandomForestClassifier(n_estimators=10, random_state=125)),
     ]
@@ -46,7 +48,7 @@ predictions = pipe.predict(X_test)
 accuracy = accuracy_score(y_test, predictions)
 f1 = f1_score(y_test, predictions, average="macro")
 
-print("Accuracy: ", str(round(accuracy, 2)*100)+"%","F1: ", round(f1, 2))
+print("Accuracy: ", str(round(accuracy, 2) * 100) + "%", "F1: ", round(f1, 2))
 
 # Confusion matrix
 import matplotlib.pyplot as plt
@@ -64,8 +66,9 @@ with open("./results/metrics.txt", "w") as outfile:
 
 # Save the model
 import pickle
+
 # save the model to disk
-filename = './model/drug_pipeline.sav'
-pickle.dump(pipe, open(filename, 'wb'))
+filename = "./model/drug_pipeline.sav"
+pickle.dump(pipe, open(filename, "wb"))
 
-#sio.dump(pipe, "./model/drug_pipeline.skops")
+# sio.dump(pipe, "./model/drug_pipeline.skops")