Getting Started With Prophet

Source

The Departure

This is a first look at Prophet a time-series forecasting library. It depends on pystan, a Bayesian modelling platform, which in turn depends on numpy and cython.

Imports

Python

from pathlib import Path
import os

PyPi

from dotenv import load_dotenv
import modin.pandas as pandas

My Stuff

from graeae.timers import Timer

Setup

The Paths

load_dotenv(".env", override=True)
DATA = Path(os.environ.get("PEYTON_MANNING_WIKIPEDIA_VIEWS")).expanduser()
assert DATA.is_file()

The Timer

TIMER = Timer()

The Initiation

Load The Data

with TIMER:
    data = pandas.read_csv(DATA)
print(data.shape)

WARNING: Falling back to serializing objects of type <class 'pathlib.PosixPath'> by using pickle. This may be inefficient.
Started: 2019-04-05 15:23:56.038686
Ended: 2019-04-05 15:23:56.236422
Elapsed: 0:00:00.197736
(2905, 2)