mlaas_train_and_visualize

def confusion_matrix_plot(data):
    z = data.tolist()[::-1]
    x = ['Negative','Positive']
    y = ['Positive','Negative']
    z_text = z

    fig = ff.create_annotated_heatmap(z, x, y, annotation_text=z_text, text=z,hoverinfo='text',colorscale='Blackbody')
    fig.update_layout(font_family="IBM Plex Sans")

    st.write(fig)

def roc_plot(data):
    fig = px.line(data, x="False Positive", y="True Positive")#, title='ROC Curve')
    fig.update_layout(font_family="IBM Plex Sans")

    st.write(fig)
    
try:
    X = df[feature_cols]
    y = df[label_col]
    X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=test_size,random_state=0)
    logreg = LogisticRegression(penalty=penalty, dual=dual, tol=tol, C=C,
                                fit_intercept=fit_intercept, intercept_scaling=intercept_scaling,
                                class_weight=class_weight, random_state=random_state, solver=solver,
                                max_iter=max_iter, multi_class=multi_class, verbose=verbose,
                                warm_start=warm_start, l1_ratio=l1_ratio)
    logreg.fit(X_train,y_train)
    y_pred = logreg.predict(X_test)
    cnf_matrix = metrics.confusion_matrix(y_test, y_pred)

    st.subheader('Confusion Matrix')
    confusion_matrix_plot(cnf_matrix)

    accuracy = metrics.accuracy_score(y_test, y_pred)
    precision = metrics.precision_score(y_test, y_pred)
    recall = metrics.recall_score(y_test, y_pred)
    f1 = metrics.f1_score(y_test, y_pred)

    st.subheader('Metrics')
    col2_1, col2_2, col2_3, col2_4 = st.columns(4)

    with col2_1:
        st.info('Accuracy: **%s**' % (round(accuracy,3)))
    with col2_2:
        st.info('Precision: **%s**' % (round(precision,3)))
    with col2_3:
        st.info('Recall: **%s**' % (round(recall,3)))
    with col2_4:
        st.info('F1 Score: **%s**' % (round(f1,3)))

    y_pred_proba = logreg.predict_proba(X_test)[::,1]
    fpr, tpr, _ = metrics.roc_curve(y_test,  y_pred_proba)
    roc_data = pd.DataFrame([])
    roc_data['True Positive'] =  tpr
    roc_data['False Positive'] = fpr
    st.subheader('ROC Curve')
    roc_plot(roc_data)
    auc = metrics.roc_auc_score(y_test, y_pred_proba)
    st.info('Area Under Curve: **%s**' % (round(auc,3)))