app.py

import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns
import streamlit as st
from sklearn.datasets import make_regression
from sklearn.metrics import mean_squared_error
from sklearn.neighbors import KNeighborsRegressor

st.subheader("K nearest neighbor (KNN) Regressor")

st_col = st.columns(1)[0]

K = st.slider(
    "Number of nearest neighbors (K)", min_value=1, max_value=10, value=5, step=1
)

option = st.selectbox(
    "Select Distance Metric", ("L1(Manhattan)", "L2(Euclidean Distance)")
)

X, y = make_regression(n_samples=100, n_features=1, noise=0.3, random_state=42)

ntrain = 30

x_train = X[:ntrain]
y_train = y[:ntrain]
x_test = X[ntrain:]
y_test = y[ntrain:]

if str(option) == "L1(Manhattan)":
    metric = "manhattan"
else:
    metric = "euclidean"

knn = KNeighborsRegressor(n_neighbors=K, metric=metric)
knn.fit(x_train, y_train)
y_pred = knn.predict(x_test)

plt.figure()

plt.plot(y_test[:30], "C0s", label="True Points (Test)")
plt.plot(y_pred[:30], "C1*", label="Predictions (Test)")
plt.xlabel("X")
plt.ylabel("Y")
plt.legend(loc="upper left")
plt.ylim(-90,90)

sns.despine(right=True, top=True)

with st_col:
    st.pyplot(plt)
error = mean_squared_error(y_test, y_pred)
st.write("The mean squared error is %.2f" % error)

hide_streamlit_style = """
            <style>
            #MainMenu {visibility: hidden;}
            footer {visibility: hidden;}
            subheader {alignment: center;}
            </style>
            """
st.markdown(hide_streamlit_style, unsafe_allow_html=True)

st.markdown(
    """
    The above plot shows the True values and Predictions for 30 points in the test set.
    It can be observed that the optimal value of K is 3 for the given dataset.
    """,
    unsafe_allow_html=True,
)