lppier

source ~/shrink_venv/bin/activate
cd $VIRTUAL_ENV/lib/python3.6/site-packages
zip -r9 ~/CreateThumbnail.zip *
cd ~
zip -g CreateThumbnail.zip CreateThumbnail.py
aws lambda update-function-code --function-name CreateThumbnail --zip-file fileb:///home/ec2-user/CreateThumbnail.zip

lppier

from datetime import date
from dateutil.relativedelta import relativedelta, MO

# Relative delta replaces the day in the date you specify by day = 31. From this new date, weekday=MO(-1) specifies the last monday. 

def get_last_mon_of_mth(dt):
    print (dt + relativedelta(day=31, weekday=MO(-1)))

get_last_mon_of_mth(date(2018, 12, 30))
get_last_mon_of_mth(date(2019, 6, 4))

lppier

from datetime import date
from dateutil.relativedelta import relativedelta, MO

# Relative delta replaces the day in the date you specify by day = 1. From this new date, weekday=MO(+1) specifies the 1st Monday. 

def get_first_mon_of_mth(dt):
    print (dt + relativedelta(day=1, weekday=MO(+1)))

get_first_mon_of_mth(date(2018, 12, 30))
get_first_mon_of_mth(date(2019, 6, 4))

lppier

import pandas as pd
import numpy as np
import redis
import pyodbc
from time import time
from datetime import date, timedelta, datetime

r = redis.Redis(host='localhost', port=6379, db=0)
dt_last = datetime.strptime(r.get("last_update_time").decode(), '%Y-%m-%d %H:%M:%S')
dt_now = datetime.now()

lppier

import redis
import json

# Retrieve Redis Stored Values, if any for faster processing
anc_dict_320 = {}
r = redis.Redis(host='localhost', port=6379, db=0)
json_str_320 = r.get("anc_dict_320")
if json_str_320 is not None:
    anc_dict_320 = json.loads(json_str_320)
    print(anc_dict_320)

lppier

import pyodbc
import pandas as pd

# https://docs.microsoft.com/en-us/sql/connect/odbc/linux-mac/installing-the-microsoft-odbc-driver-for-sql-server?view=sql-server-2017

conn = pyodbc.connect(
    r'DRIVER={ODBC Driver 17 for SQL Server};'
    r'SERVER=mydbinstance.cazdl38iyoz0.ap-southeast-1.rds.amazonaws.com;'
    r'DATABASE=testdb;'
    r'UID=pier;'

lppier

import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
import matplotlib.style as style

style.use('ggplot')

f, ax1 = plt.subplots(1, 1, figsize=(20, 7), sharex=True)
df_mth_plot = pd.DataFrame(list(monthly_growth_dict.items()), columns=['Market', 'Average Monthly Growth Rate %'])
df_mth_plot = df_mth_plot.sort_values('Average Monthly Growth Rate %', ascending=False)

lppier

import calendar
df_weekly_season['DOW_ENG'] = weekly_season['ds'].apply(lambda x : calendar.day_name[pd.to_datetime(x).weekday()])
df_weekly_season['DOW'] = weekly_season['ds'].apply(lambda x : int(pd.to_datetime(x).strftime('%w'))) # SUN == 0
df_weekly_season['DOW'] = weekly_season['ds'].apply(lambda x : int(pd.to_datetime(x).weekday()) # MON == 0

lppier

> yum install centos-release-scl
> yum info rh-python36
> scl enable rh-python36 bash
> python --version

Run PySpark2

> pyspark2

Python 3.6.3 (default, Apr 26 2018, 13:16:02) 

lppier

#SettingWithCopyWarning: 
#A value is trying to be set on a copy of a slice from a DataFrame.
#Try using .loc[row_indexer,col_indexer] = value instead

# assign was introduced in pandas 0.16 to deal with this false positive

# Instead of 
df_weekly_season.loc[:, 'market'] = market 
# or
df_weekly_season['market'] = market