Upload 2 files

requirements.txt

@@ -1,5 +1,5 @@
-polars
-pandas
-streamlit
-scikit-learn
-plotly
+pandas
+polars
+streamlit
+plotly
+openmeteo-requests

streamlit_app.py

@@ -0,0 +1,195 @@
+from collections import namedtuple
+import altair as alt
+import math
+import pandas as pd
+import streamlit as st
+import polars as pl
+import openmeteo_requests
+from datetime import *
+
+st.set_page_config(
+    page_title="Weather Data",
+    page_icon="🌤️",
+	layout='wide'
+)
+
+st.markdown("# Historic Weather Data for 3 BYU Campuses")
+st.markdown("__Explore weather data gathered at all 3 BYU locations from any chosen date range. Observe visuals explaining the temperature, pressure, wind speed, and far more from all locations.__")
+
+columns = st.columns(3, gap='medium')
+
+openmeteo = openmeteo_requests.Client()
+
+st.sidebar.markdown("_Note: Data is only present up to 4 days ago_")
+
+length = st.sidebar.slider("Length of Time Period for Data (Days)", min_value=5, max_value=30, value=15)
+
+campus = ['provo', 'hawaii', 'idaho']
+locations = [(40.2518, 111.6493), (21.6419, 157.9267), (43.8145, 111.7833)]
+
+start_day = st.sidebar.date_input("Start of Period for Data", value=(datetime.today() - timedelta(days=35)), min_value=date(2020, 1, 1), max_value=(datetime.today() - timedelta(days=(length+4))))
+
+end = (start_day + timedelta(days=length)).strftime('%Y-%m-%d')
+start = start_day.strftime('%Y-%m-%d')
+
+zone_select = st.sidebar.selectbox("Choose Timezone", options=['Mountain Time', 'Hawaiian Time'])
+
+zone = 'HST' if zone_select == 'Hawaiian Time' else 'MST'
+
+variables = ['Temperature (F)', 'Apparent Temp (F)', 'Humidity (%)', 'Pressure (hPa)', 'Precipitation (in)', 'Rain (in)', 'Snowfall (in)', 'Cloud Cover (%)', 'Wind Speed (mph)', 'Snow Depth (m)']
+variable_select = st.sidebar.selectbox("Choose variable of interest", options=variables)
+variable_dict = {
+	'Temperature (F)': 'temperature_2m',
+	'Apparent Temp (F)': 'apparent_temperature',
+	'Humidity (%)': 'relative_humidity_2m',
+	'Pressure (hPa)': 'surface_pressure',
+	'Precipitation (in)': 'precipitation',
+	'Rain (in)': 'rain',
+	'Snowfall (in)': 'snowfall',
+	'Cloud Cover (%)': 'cloud_cover',
+	'Wind Speed (mph)': 'wind_speed_10m',
+	'Snow Depth (m)': 'snow_depth'
+}
+
+var = variable_dict[variable_select]
+
+hi_lo = st.sidebar.selectbox("Daily Extremes", options=['Max', 'Min'])
+
+daily_var = 'temperature_2m_max' if hi_lo == 'Max' else 'temperature_2m_min'
+
+def request(url, lat, long, start, end, variable, time_unit, timezone):
+    params = {
+	"latitude": lat,
+	"longitude": long,
+	"start_date": start,
+	"end_date": end,
+	time_unit: variable,
+    "temperature_unit": "fahrenheit",
+	'timezone': timezone,
+	'precipitation_unit': 'inch',
+	'wind_speed_unit': 'mph'
+	}
+    return openmeteo.weather_api(url, params=params)[0]
+
+i = 0
+temp_box = pl.DataFrame()
+for location in locations:
+	actual = request("https://archive-api.open-meteo.com/v1/archive", location[0], location[1], start, end, var, 'hourly', zone)
+	forecasted = request("https://historical-forecast-api.open-meteo.com/v1/forecast", location[0], location[1], start, end, var, 'hourly', zone)
+
+	hourly_actual= actual.Hourly()
+	hourly_var_actual = hourly_actual.Variables(0).ValuesAsNumpy()
+
+	hourly_forecasted = forecasted.Hourly()
+	hourly_var_forecasted = hourly_forecasted.Variables(0).ValuesAsNumpy()
+
+	hourly_data = {"date-time": pd.date_range(
+		start = pd.to_datetime(hourly_actual.Time(), unit = "s", utc = True),
+		end = pd.to_datetime(hourly_actual.TimeEnd(), unit = "s", utc = True),
+		freq = pd.Timedelta(seconds = hourly_actual.Interval()),
+		inclusive = "left"
+	)}
+
+	hourly_data[var] = hourly_var_actual
+	hourly_data[f"forecast_{var}"] = hourly_var_forecasted
+
+	hourly_dataframe = pl.DataFrame(hourly_data)
+
+	hourly_dataframe = hourly_dataframe.with_columns(pl.col('date-time').dt.strftime('%Y-%m-%d').alias('date'))
+	hourly_dataframe = hourly_dataframe.with_columns(pl.col('date-time').dt.strftime("%H:%M:%S").alias('time'))
+
+	with columns[i]:
+		st.markdown(f"### {campus[i].title()}")
+		st.dataframe(hourly_dataframe.drop('date-time'))
+		st.dataframe(hourly_dataframe.select([var, f'forecast_{var}']).describe()[2:9])
+      
+	current_campus = hourly_dataframe.with_columns(pl.lit(campus[i]).alias('campus'))
+
+	end_month = (datetime.today() - timedelta(days=4)).strftime('%Y-%m-%d')
+	start_month = (datetime.today() - timedelta(days=35)).strftime('%Y-%m-%d')
+	daily = request("https://archive-api.open-meteo.com/v1/archive", location[0], location[1], start_month, end_month, daily_var, 'daily', zone).Daily()
+
+	daily_data = {"date-time": pd.date_range(
+		start = pd.to_datetime(daily.Time(), unit = "s", utc = True),
+		end = pd.to_datetime(daily.TimeEnd(), unit = "s", utc = True),
+		freq = pd.Timedelta(seconds = daily.Interval()),
+		inclusive = "left"
+	)}
+	daily_data['temperature'] = daily.Variables(0).ValuesAsNumpy()
+
+	daily_df = pl.DataFrame(daily_data)
+	daily_df = daily_df.with_columns(pl.lit(campus[i]).alias('campus'))
+
+	if len(temp_box) == 0:
+		temp_box = pl.DataFrame(current_campus)
+		temp_line = daily_df
+	else:
+		temp_box = temp_box.vstack(current_campus)
+		temp_line = temp_line.vstack(daily_df)
+
+	i+=1
+
+daily = alt.Chart(temp_line).mark_line().encode(
+	alt.X("date-time"),
+	alt.Y('temperature'),
+	color='campus'
+).properties(
+    title='Daily Temp Extremes Over Last Month',
+	width=300,
+	height=300
+)
+
+
+hourly_box = alt.Chart(temp_box).mark_boxplot().encode(
+	alt.X(var).scale(zero=False),
+	alt.Y("campus")
+).properties(
+    title='Hourly Readings',
+	width=400,
+	height=300
+)
+
+
+grouped = temp_box.select(['time', var, 'campus']).group_by(['time', 'campus']).mean()
+
+hourly_average = alt.Chart(grouped).mark_line().encode(
+	x="time",
+	y=var,
+	color="campus"
+).properties(
+    title='Average Reading Given Time of Day',
+	width=400,
+	height=300
+)
+
+
+kpi_max = temp_box.select([var, 'campus']).group_by('campus').max()
+
+pr_hi = kpi_max.filter(pl.col('campus') == 'provo')[var][0]
+id_hi = kpi_max.filter(pl.col('campus') == 'idaho')[var][0]
+hi_hi = kpi_max.filter(pl.col('campus') == 'hawaii')[var][0]
+
+
+kpi_low = temp_box.select([var, 'campus']).group_by('campus').min()
+pr_lo = kpi_low.filter(pl.col('campus') == 'provo')[var][0]
+id_lo = kpi_low.filter(pl.col('campus') == 'idaho')[var][0]
+hi_lo = kpi_low.filter(pl.col('campus') == 'hawaii')[var][0]
+
+
+st.sidebar.altair_chart(daily)
+
+columns[1].altair_chart(hourly_box)
+
+columns[0].altair_chart(hourly_average)
+
+with columns [2]:
+	sub_cols = st.columns(2)
+	with sub_cols[0]:
+		st.markdown(f"Provo Max: {round(pr_hi, 1)}")
+		st.markdown(f"Hawaii Max: {round(hi_hi, 1)}")
+		st.markdown(f"Idaho Max: {round(id_hi, 1)}")
+	
+	with sub_cols[1]:
+		st.markdown(f"Provo Min: {round(pr_lo, 1)}")
+		st.markdown(f"Hawaii Min: {round(hi_lo, 1)}")
+		st.markdown(f"Idaho Min: {round(id_lo, 1)}")