Data_Visualization_With_Streamlit.py

import streamlit as st
import pandas as pd  # type: ignore
import numpy as np
import pydeck as pdk  # type: ignore
import os
import matplotlib.pyplot as plt  # type: ignore
from dotenv import load_dotenv  # type: ignore

# get url as environment var
load_dotenv()
url = str(os.environ.get("URL"))
path = str(os.environ.get("FILEPATH"))

st.set_page_config(
    page_title="Data Visualization With Streamlit",
    page_icon="📈",
)

types = {"latitude": np.float32, "longitude": np.float32}

st.title("Houses for Sale for loft.com.br")
st.subheader("Jardim América, São Paulo, Brasil")
st.markdown(
    """
This app shows a small demo of data taken by web scraping of a brazilian estate website (loft.br)
and some data visualization about the data retrieved."""
)
st.markdown(
    """**In this page**, you can see data visualization 
graphics done with python (package matplotlib) from a default "houses_df_original.csv" or your own "houses_df.csv
from your web scraping."""
)

uploaded_file = "None"
with st.expander("Upload your own 'houses_df.csv' file"):
    uploaded_file = str(st.file_uploader("Upload file"))
if (uploaded_file == "None") and (os.path.exists(path)):
    uploaded_file = path
if uploaded_file != "None":
    # Can be used wherever a "file-like" object is accepted:
    df = pd.read_csv(uploaded_file)
if len(df) == 0:
    st.write("No data to visualize")
else:
    for col in df.columns:
        if "Unnamed" in col:
            df = df.drop(columns=col)
            break

    options = st.multiselect(
        "Select one or more house types:",
        df.house_type.unique(),
        [
            df.house_type.unique()[0],
            df.house_type.unique()[1],
            df.house_type.unique()[2],
        ],
    )

    df_selected = df[df["house_type"].isin(options)]

    df_nums = df_selected.drop(
        columns=["house_links", "address", "latitude", "longitude"]
    )
    df_nums = df_nums.groupby("house_type").agg(["mean", "max", "min"]).reset_index()
    df_nums.columns = [
        "House Type",
        "House Price Average",
        "House Price Max",
        "House Price Min",
        "Area Average",
        "Area Max",
        "Area Min",
        "Bedrooms Average",
        "Bedrooms Max",
        "Bedrooms Min",
        "Parking Space Average",
        "Parking Space Max",
        "Parking Space Min",
    ]  # type: ignore
    df_nums["Number of Houses"] = (
        df_selected.groupby("house_type").count().reset_index()["house_links"]
    )
    df_nums["House Price per m2"] = (
        df_nums["House Price Average"] / df_nums["Area Average"]
    )

    # Define a layer to display on a map
    layer = pdk.Layer(
        "ScatterplotLayer",
        df_selected,
        pickable=True,
        opacity=0.8,
        stroked=True,
        filled=True,
        auto_highlight=True,
        extruded=True,
        elevation_scale=0.1,
        elevation_range=[0, 1],
        radius=1,
        radius_scale=1,
        radius_min_pixels=1,
        radius_max_pixels=100,
        line_width_min_pixels=1,
        get_position=["longitude", "latitude"],
        get_radius=8,
        get_fill_color=[102, 179, 72],
        get_line_color=[0, 0, 0],
    )

    # Set the viewport location
    view_state = pdk.ViewState(
        latitude=-23.562799, longitude=-46.663020, zoom=14, bearing=0, pitch=0
    )

    # Render
    r = pdk.Deck(
        layers=[layer],
        initial_view_state=view_state,
        map_style="road",
        tooltip={"text": "{name}\n{address}"},
    )

    col1, col2 = st.columns([1, 2])
    with col1:
        st.metric(
            label="Number of Houses", value=f'{df_nums["Number of Houses"].sum():,}'
        )
        st.metric(label="Max. Area", value=f'{df_nums["Area Max"].max():,} m2')
        st.metric(label="Min. Area", value=f'{df_nums["Area Min"].min():,} m2')
        st.metric(
            label="Max. House Price", value=f'${df_nums["House Price Max"].max():,}'
        )
        st.metric(
            label="Min. House Price", value=f'${df_nums["House Price Min"].min():,}'
        )
    with col2:
        st.pydeck_chart(r, use_container_width=False)

    tab1, tab2 = st.tabs(["📈 Charts", "🗃 Data"])

    col1, col2 = tab1.columns(2)

    with col1:
        st.bar_chart(df_nums, y="House Price Average", x="House Type")
        st.bar_chart(df_nums, y="House Price per m2", x="House Type")

    with col2:
        st.bar_chart(df_nums, y="Area Average", x="House Type")
        plt.pie(
            df_nums["Number of Houses"], labels=df_nums["House Type"], autopct="%.0f%%"
        )
        st.pyplot(plt)  # type: ignore[arg-type]

    tab2.dataframe(df_nums.set_index("House Type").T)