covid_world.py

import numpy as np
import pandas as pd
import seaborn as sns
import streamlit as st


# Youtube Tutorial
# https://www.youtube.com/channel/UC3LD42rjj-Owtxsa6PwGU5Q

@st.cache
def load_data(data_path):
    df = pd.read_csv(data_path)
    df["date"] = pd.to_datetime(df["date"])
    df.set_index("date", inplace=True)
    df["cases"] = df.groupby("location")["new_cases"].transform(
        lambda x: x.rolling(7).mean()
    )
    df["deaths"] = df.groupby("location")["new_deaths"].transform(
        lambda x: x.rolling(7).mean()
    )
    return df


df = load_data("covid_world_data.csv")

"## Display Data"
df

df_country_tot = df.groupby("location")[["new_cases", "new_deaths"]].sum()
df_country_tot["mortality_ratio"] = df_country_tot["new_deaths"] / (
    df_country_tot["new_deaths"] + df_country_tot["new_cases"]
)
"## Top 10 Countries by Mortality Ratio"
st.write(df_country_tot.sort_values("mortality_ratio", ascending=False).head(10))

"## Top 10 Countries by cases"
st.write(df_country_tot.sort_values("new_cases", ascending=False).head(10))

country = st.sidebar.selectbox("Country: ", df["location"].unique())
st.write(f"### {country} Trend")
df[df["location"] == country][["cases", "deaths"]].plot()
st.pyplot()

countries = st.multiselect("Select countries", df["location"].unique())
choice_col = st.radio("Compare: ", ("deaths", "cases"))
df[df["location"].isin(countries)].groupby("location")[choice_col].plot(legend=True)
st.pyplot()