GDBSD

def compare_table_schemas(client, project: str, dataset: str, 
                          table_a: str, table_b: str) -> bool:
    """Compare the schemas of two BigQuery tables. Useful for instance
    to confirm that there hasn't been a drift in the schemas for the
    production and development tables.
    
    :param client: BigQuery client object
    :param project: string - GCP project ID
    :param dataset: string - dataset name
    :param table_a:  string - table name

GDBSD

def calc_beta_f1(beta, precision, recall):
    """Calculate F-beta given the values for precision and recall and the beta value"""
    beta_sq = pow(beta, 2)
    num = (1 + beta_sq)*precision*recall
    denom = beta_sq*precision+recall
    return num/denom

GDBSD

def remove_non_printing_chars(df):
    """Clean a dataframe column to remove any non-printing characters.
    We've encountered values like tabs in some of the data.

    :param df: Pandas dataframe
    :return: Pandas dataframe
    """
    clean_df = df.copy(deep=True)
    clean_df = clean_df.apply(lambda x: x.str.strip() if x.dtype == "object" else x)
    for col in list(clean_df.columns):

GDBSD

import os
import sys

"""Utility function to update sys.path so in our notebooks you can import modules from 
any folder in the application. It will also allow you to import any module in your virtual
environment. Note that in my project the virtual environment is named "venv".

In the notebook, in the first cell, import this script. It will run
automatically

GDBSD

# Use Case: Here we have a dict "stats" with four keys with arrays and floats as values, both of which can trip you up.
# We solve it by using Numpy .all() and PyTest approx()

    assert type(stats['observed']) is np.ndarray
    assert type(stats['expected']) is np.ndarray
    assert (stats['observed'] == [[1, 2, 3], [4, 5, 6], [7, 8, 9]]).all()
    assert (stats['expected'] == [[1.6, 2.0, 2.4], [4.0, 5.0, 6.0], [6.4, 8.0, 9.6]]).all()
    assert stats['G'] == pytest.approx(0.49173089057312613)
    assert stats['p'] == pytest.approx(0.9743009689044624)

GDBSD

# Consider this code
x = 5 
def func1(): 
    print(x) 
func1() 
# Output
5

# Since x is declared before the function call, func1 can access it. 
# However, if you try to change it:

GDBSD

import numpy as np

def test_dict_equality(dict_1, dict_2):
    false_matches = 0
    for key in dict_1:
        if key in dict_2:
            if not np.array_equal(dict_2[key], dict_2[key]):
                false_matches += 1
    return false_matches == 0

GDBSD

# I've seen a lot of posts with instructions for opening a Jupyter Notebook on your
# local device with the Juyter server running on a GCP VM. They make it seem really
# complicated. It ain't that hard folks!

1. On the GCP Compute Engine UI click on the drop-down menu on the upper left side 
   under Remote access. Select "view gcloud command" and copy the command.

2. To that command append   -- -L localhost:8887:127.0.0.1:8889 
     Example: 
     gcloud beta compute ssh --zone "<zone-name>" "<vm-instanve-name>" --project "<project-name>"  -- -L localhost:8887:127.0.0.1:8889

GDBSD

import gzip
import json


source_dict = {
    "New Year's Day": "Fri, Jan 1, 2021",
    "Martin Luther King Jr. Day": "Mon, Jan 18, 2021",
    "Washington's Birthday": "Mon, Feb 15, 2021",
    "Arbor Day": "Fri, Apr 30, 2021",
    "Memorial Day": "Mon, May 31, 2021",