Skip to content

Commit 77b4091

Browse files
IntegerManIntegerMan
committedJun 19, 2022
Ready for rehearsal
1 parent ea7d332 commit 77b4091

10 files changed

+128155
-78850
lines changed
 

‎data/matchup_predictions.csv

+992-930
Large diffs are not rendered by default.

‎part0_install.ipynb

-141
This file was deleted.

‎part0_intro.ipynb

+120
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,120 @@
1+
{
2+
"cells": [
3+
{
4+
"cell_type": "markdown",
5+
"metadata": {},
6+
"source": [
7+
"# Automating Machine Learning with Python and Azure\n",
8+
"By [Matt Eland](https://MattEland.dev) | [@IntegerMan](https://twitter.com/IntegerMan)"
9+
]
10+
},
11+
{
12+
"cell_type": "markdown",
13+
"metadata": {},
14+
"source": [
15+
"## Who am I?\n",
16+
"\n",
17+
"After 20+ years of software engineering and engineering management I now teach professionally and do many things including: \n",
18+
"\n",
19+
"- Teaching Software Development @ [Tech Elevator](https://TechElevator.com) in Columbus, OH\n",
20+
"- Speaking at [Conference & User Groups](https://sessionize.com/matt-eland)\n",
21+
"- Pursuing a Master's Degree in [Data Analytics through Franklin University](https://www.franklin.edu/degrees/masters/data-analytics) _(Anticipated Graduation: Summer of 2024)_\n",
22+
"- Data Science experiments for fun!\n",
23+
"- Data Science blogging at [AccessibleAI.dev](https://accessibleai.dev)\n",
24+
"- Data Science YouTube content at [MattOnDataScience.com](https://MattOnDataScience.com)\n",
25+
"- Co-organizing the [Central Ohio .NET Developer Group](https://condg.org)\n",
26+
"\n",
27+
"I also have several additional qualifications that relate to this talk:\n",
28+
"\n",
29+
"- [Microsoft Certified: Azure Data Scientist Associate](https://www.credly.com/badges/53ace869-2160-4fcd-be00-271bc5ada4aa/public_url)\n",
30+
"- [Microsoft Certified: Azure AI Fundamentals](https://www.credly.com/badges/bf234824-94ff-4d28-8080-cf010ff5408e/public_url)\n",
31+
"- [IBM Data Scientist Professional Certificate](https://www.credly.com/earner/earned/badge/e2625f64-3fe5-4028-9db2-2a28dabf6405)\n",
32+
"- [Machine Learning Engineer with Microsoft Azure Nanodegree](https://confirm.udacity.com/DWWTFM2U)\n"
33+
]
34+
},
35+
{
36+
"cell_type": "markdown",
37+
"metadata": {},
38+
"source": [
39+
"## What's this talk about?\n",
40+
"\n",
41+
"**Machine Learning in Azure using Python**. Seems kinda obvious.\n",
42+
"\n",
43+
"Some specific areas we'll cover:\n",
44+
"\n",
45+
"- What is Azure Machine Learning Studio?\n",
46+
"- What is the Azure ML SDK?\n",
47+
"- Why would I even want to use Azure for this?\n",
48+
"- What does Python code look like for working with Azure?\n",
49+
"\n",
50+
"For the bulk of this talk, we'll explore a **regression experiment** on hockey penalty prediction from data exploration to automated machine learning to model deployment and consumption.\n",
51+
"\n",
52+
"We'll also look briefly at a prior **classification experiment** on Die Hard to see the types of metrics available when looking at classification experiments."
53+
]
54+
},
55+
{
56+
"cell_type": "markdown",
57+
"metadata": {},
58+
"source": [
59+
"## Touring Azure Machine Learning Studio\n",
60+
"\n",
61+
"Let's take a look at Azure Machine Learning Studio to see what the Azure ML SDK allows us to automate.\n",
62+
"\n",
63+
"![](screens/AutoMLModelPerformance.png)"
64+
]
65+
},
66+
{
67+
"cell_type": "markdown",
68+
"metadata": {},
69+
"source": [
70+
"## What is the Azure ML SDK?\n",
71+
"\n",
72+
"A way of interacting with the Azure ML SDK from Python code.\n",
73+
"\n",
74+
"It allows you to largely bypass the Azure Portal and can integrate more easily into other workflows.\n",
75+
"\n",
76+
"To install the Azure ML SDK, you should follow [Microsoft's official documentation](https://docs.microsoft.com/en-us/python/api/overview/azure/ml/installv2?view=azure-ml-py) as the ML SDK continues to evolve."
77+
]
78+
},
79+
{
80+
"cell_type": "markdown",
81+
"metadata": {},
82+
"source": [
83+
"## When would I use Azure ML?\n",
84+
"\n",
85+
"- When you don't know which algorithms to evaluate\n",
86+
"- When you want to avoid or defer learning curves around data science libraries\n",
87+
"- When you want easy access to advanced metrics, explanations, and visualizations\n",
88+
"- When you want to version your datasets and models in the cloud\n",
89+
"- When you want to be able to easily share the results of experiments \n",
90+
"- When you need extra computing power for ML paid at a usage level\n",
91+
"- When you want to deploy, monitor, and scale your trained models"
92+
]
93+
}
94+
],
95+
"metadata": {
96+
"interpreter": {
97+
"hash": "a4868653bb6f8972e87e4c446ab8a445a15b25dedb8594cc74c480f8152ea86a"
98+
},
99+
"kernelspec": {
100+
"display_name": "Python 3.8.8 64-bit",
101+
"language": "python",
102+
"name": "python3"
103+
},
104+
"language_info": {
105+
"codemirror_mode": {
106+
"name": "ipython",
107+
"version": 3
108+
},
109+
"file_extension": ".py",
110+
"mimetype": "text/x-python",
111+
"name": "python",
112+
"nbconvert_exporter": "python",
113+
"pygments_lexer": "ipython3",
114+
"version": "3.8.8"
115+
},
116+
"orig_nbformat": 4
117+
},
118+
"nbformat": 4,
119+
"nbformat_minor": 2
120+
}

0 commit comments

Comments
 (0)