-
Notifications
You must be signed in to change notification settings - Fork 3
/
app.py
131 lines (110 loc) · 5.63 KB
/
app.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
import os
import pytz
import boto3
import pickle
import datetime
import pandas as pd
import streamlit as st
import streamlit_analytics
from src.visualize_data import avg_data_day, given_day, preprocess_current_data, plot_data
bucketname = 'boulderbucket'
dfname = 'boulderdata.csv'
modelname = 'model.dat'
gymdataname = 'gymdata.json'
s3 = boto3.client('s3',
aws_access_key_id=st.secrets.aws_access_key_id,
aws_secret_access_key=st.secrets.aws_secret_access_key,
region_name=st.secrets.region
)
# Download files from S3
s3.download_file(bucketname, dfname, dfname)
s3.download_file(bucketname, gymdataname, gymdataname)
if not os.path.isfile('firestore-key.json'):
s3.download_file(bucketname, 'firestore-key.json', 'firestore-key.json')
def st_given_day(boulderdf, gymdatadf):
# ask user for gym and date input
st.markdown("## How full is my gym today?")
selected_gym = st.radio('Select a gym', sorted(list(boulderdf['gym_name'].unique())))
today = datetime.date.today()
# first available date is one month before the current date
first_date = today - datetime.timedelta(days=30)
selected_date = st.date_input('Selected date', today, min_value=first_date, max_value=today)
st.markdown(f"### Showing results for [{selected_gym}]({gymdatadf[selected_gym]['url']})\nOccupancy shown in percentage")
# display the data for the given day
givendaydf = given_day(boulderdf, str(selected_date), selected_gym)
if givendaydf.empty:
st.error('There is no data to show for this day. The gym might be closed')
else:
st.altair_chart(plot_data(givendaydf), use_container_width=True)
return selected_gym, selected_date
def get_current_time():
# https://stackoverflow.com/a/60169568/4569908
dt = datetime.datetime.now()
timeZone = pytz.timezone("Europe/Berlin")
aware_dt = timeZone.localize(dt)
if aware_dt.dst() != datetime.timedelta(0,0):
#summer time
dt += datetime.timedelta(hours=2)
else:
#winter time
dt += datetime.timedelta(hours=1)
dt = dt.strftime("%Y/%m/%d %H:%M")
#round to the nearest 20min interval
minutes = [0, 20, 40]
current_min = int(dt.split(':')[1])
distances = [abs(current_min - _min) for _min in minutes]
closest_min = minutes[distances.index(min(distances))]
current_time = dt.replace(':'+str(current_min), ':'+str(closest_min))
current_time = datetime.datetime.strptime(current_time, "%Y/%m/%d %H:%M")
return current_time
def st_prediction(boulderdf, selected_gym):
# prediction for future
s3.download_file(bucketname, modelname, modelname)
model = pickle.load(open(modelname, "rb"))
current_time = get_current_time()
X_today = preprocess_current_data(boulderdf, selected_gym, current_time)
prediction = int(model.predict(X_today)[0])
# get time for next interval. round to the nearest 20min
next_min = (current_time + (datetime.datetime.min - current_time) % datetime.timedelta(minutes=20)).strftime('%H:%M')
st.markdown(f'**We predict the occupancy at {next_min} will be: {prediction}**\n')
return
def st_avg_data(boulderdf, selected_gym):
st.markdown(f"""
## Average data for {selected_gym}\n
This plot shows the average occupancy, queue and weather for the given weekday.
""")
today = datetime.date.today()
weekdays = [(today + datetime.timedelta(days=x)).strftime("%A") for x in range(7)]
avg_day = st.radio('Select day of the week', weekdays)
avgdf = avg_data_day(boulderdf, weekdays.index(avg_day), selected_gym)
if avgdf.empty:
st.error('There is no data to show at all. The gym might be closed for a long time')
else:
st.altair_chart(plot_data(avgdf), use_container_width=True)
return
if __name__ == "__main__":
gymdatadf = pd.read_json(gymdataname)
st.set_page_config(
page_title="Bouldern",
#layout="wide",
page_icon="https://emojipedia-us.s3.dualstack.us-west-1.amazonaws.com/thumbs/120/apple/271/person-climbing_1f9d7.png")
st.title('Boulder gym tracker 🧗🧗♂️🧗♀️')
st.markdown(f'**Update 06.08.2022: currently supporting {len(gymdatadf.columns)} gyms!** 🚀')
st.write("Github repo: [](https://gitHub.com/anebz/boulder)")
st.image('https://land8.com/wp-content/uploads/2017/07/Bouldering1.jpg', width=700)
# plot map with gym coordinates
st.map(pd.DataFrame([[gymdatadf[gym_name]['lat'], gymdatadf[gym_name]['long']] for gym_name in gymdatadf], columns=['lat', 'lon']))
# set up analytics
streamlit_analytics.start_tracking(firestore_key_file="firestore-key.json", firestore_collection_name="counts")
boulderdf = pd.read_csv(dfname)
selected_gym, selected_date = st_given_day(boulderdf, gymdatadf)
# only show prediction for current day
#if str(selected_date) == str(datetime.datetime.today().strftime('%Y-%m-%d')):
# st_prediction(boulderdf, selected_gym)
st_avg_data(boulderdf, selected_gym)
st.markdown(f"""
Does your gym show this occupancy data? Make a PR yourself or let us know and we'll add your gym 😎\n
Created by [anebz](https://github.com/anebz) and [AnglinaBhambra](https://github.com/AnglinaBhambra).\n
Follow us! [](https://www.twitter.com/anebzt)
[](https://www.twitter.com/_AnglinaB)""")
streamlit_analytics.stop_tracking(firestore_key_file="firestore-key.json", firestore_collection_name="counts")