veyepar: An Only One Step Ahead Guide for Machine Learning Projects

Author(s):	Chang Hsin Lee
Location	Cartoon 1
Date	jul Sat 28	Days Raw Files
Start	16:30	First Raw Start	16:13
Duration	0:30:0	Offset	0:16:09
End	17:00	Last Raw End	17:13
Chapters	00:00 0:13:43
Total cuts_time	32 min.

https://pyohio.org/2018/schedule/presentation/43/
raw-playlist raw-mp4-playlist encoded-files-playlist host archive tweet mp4 svg png

assets release.pdf An_Only_One_Step_Ahead_Guide_for_Machine_Learning_Projects.json
logs
Admin: episode episode list cut list raw files day marks day marks day image_files

State:

Locked:

clear this to unlock

Locked by:

user/process that locked.

Start:

initially scheduled time from master, adjusted to match reality

Duration:

length in hh:mm:ss

Name:

Video Title (shows in video search results)

Emails:

email(s) of the presenter(s)

Released:

has someone authorised pubication

Normalise:

Channelcopy:

m=mono, 01=copy left to right, 10=right to left, 00=ignore.

Thumbnail:

filename.png

Description:

markdown

## Abstract

What does a data scientist’s day look like? On the one hand, people say that a data scientist's day is 5% modeling and 95% cleaning data and other stuff. On the other hand, there are many more machine learning tutorials and blog posts on modeling than posts on the "other stuff" when I search online. There seems to be a lack of guidance for junior data scientists when they enter into the field who are trying to complete their first few projects.

In the last few years, I have worked on several data science projects like this, where the path to success is unclear and the journey is full of pitfalls. In this talk, I will provide practical tips on machine learning projects that I learned the hard way. I will give you 2-3 tips with examples in each stage of a machine learning project --- before, during, and after --- that will help junior data scientists or anyone working on a machine learning project navigate through the muddy data waters better.

## Outline

There are a few stages of a machine learning (ML) project, and I will give a few tips for each.

**Before ML** (7-8 minutes)

What kind of questions should I ask to the most out of the preparation stage of a machine learning project?

**Starting ML** (8-10 minutes)

How do I define success and how do I get there? What kind of model should I pick? What are some Python tools that can help me work through a project?

**Pitfalls** (8-10 minutes)

I will share examples and stories of pitalls I saw or fell into in my past projects that I wasn't aware of at the time.

Comment:

production notes