- How to Validate your Model & Data and Easily Avoid Common ML Pitfalls

How to Validate your Model & Data and Easily Avoid Common ML Pitfalls

Building good and stable ML models is hard. The wide variety of challenges in the process includes biases when building or splitting the datasets, data leakages, data quality, and integrity issues, drifts, model performance stability, and many more.

In this session we’ll explore these types of c...
Building good and stable ML models is hard. The wide variety of challenges in the process includes biases when building or splitting the datasets, data leakages, data quality, and integrity issues, drifts, model performance stability, and many more.

In this session we’ll explore these types of challenges, give real-life examples of such faults, and suggest a structure for building tests for these types of issues, to enable validating them efficiently. We’ll include a hands-on demonstration of running validation tests during the ML research phase (which you can follow along by running it locally). By the end of this session, you’ll have the knowledge about which issues to look out for in order to avoid critical problems, along with the tools for how to do so efficiently.

Table of Content:
00:00 Introduction
03:15 ML Failures and Motivation
18:10 ML Validation/Testing
24:35 Deepcheck Packages
29:15 Live Code Example
46:32 QnA

--
Download Live Code: https://bit.ly/3Of8am0
Deepchecks Github: https://github.com/deepchecks/deepchecks
Deepchecks Docs: https://docs.deepchecks.com/
Deepchecks Checks Demo: https://checks-demo.deepchecks.com/
Deepcheck Slack Community: https://www.deepchecks.com/slack
Deepchecks Website: https://www.deepchecks.com/
--
Stay tuned for upcoming events: https://online.datasciencedojo.com/ev...
Some more free tutorials: https://online.datasciencedojo.com/ca...
--
About Data Science Dojo: We offer the most trusted training to help you succeed in the world of data science. More information on the courses: https://datasciencedojo.com/
--
Like us: https://www.facebook.com/datasciencedojo
Follow us:
Connect with us: https://www.linkedin.com/company/data...
--
Also, find us on Instagram: https://instagram.com/data_science_do...
Vimeo: https://vimeo.com/datasciencedojo
--
Subscribe to our newsletter for data science content & infographics: https://datasciencedojo.com/newsletter/

#modelvalidation #machinelearning

#model validation #model validation techniques #machine learning #machine learning pitfalls #machine learning model deployment #data leakages #data quality #data integrity #model performance stability #validation tests

Data Science Dojo

🎉 90,000 人達成!  📈 予測:10万人まであと336日(2023年11月8日) 

Timetable

動画タイムテーブル

動画数:403件

Introduction - Writing Unit Tests for Data Science Code

Introduction

Writing Unit Tests for Data Science Code
2022年12月01日
00:00:00 - 00:02:16
Unit Testing in Data Science vs Dev - Writing Unit Tests for Data Science Code

Unit Testing in Data Science vs Dev

Writing Unit Tests for Data Science Code
2022年12月01日
00:02:16 - 00:03:46
Real-World Applications - Writing Unit Tests for Data Science Code

Real-World Applications

Writing Unit Tests for Data Science Code
2022年12月01日
00:03:46 - 00:08:35
Use Case and How To - Writing Unit Tests for Data Science Code

Use Case and How To

Writing Unit Tests for Data Science Code
2022年12月01日
00:08:35 - 00:32:12
Key Takeaways - Writing Unit Tests for Data Science Code

Key Takeaways

Writing Unit Tests for Data Science Code
2022年12月01日
00:32:12 - 00:33:23
Tools That Help - Writing Unit Tests for Data Science Code

Tools That Help

Writing Unit Tests for Data Science Code
2022年12月01日
00:33:23 - 00:33:40
QnA - Writing Unit Tests for Data Science Code

QnA

Writing Unit Tests for Data Science Code
2022年12月01日
00:33:40 - 00:45:16
Traffic Crashes Statistics - Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety

Traffic Crashes Statistics

Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety
2022年11月30日
00:00:00 - 00:06:31
Traffic Enforcement Systems and Technologies - Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety

Traffic Enforcement Systems and Technologies

Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety
2022年11月30日
00:06:31 - 00:10:02
Video Analytics at the Edge13:!3 Intro to Deep Neural Networks - Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety

Video Analytics at the Edge13:!3 Intro to Deep Neural Networks

Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety
2022年11月30日
00:10:02 - 00:24:12
Intelligent Traffic Monitoring - Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety

Intelligent Traffic Monitoring

Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety
2022年11月30日
00:24:12 - 00:27:28
Challenges - Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety

Challenges

Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety
2022年11月30日
00:27:28 - 00:37:00
QnA - Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety

QnA

Saving Lives Behind the Wheel: Artificial Intelligence and Computer Vision for Road Safety
2022年11月30日
00:37:00 - 00:49:42
What is data science and data science pipeline - From Data to Dashboard: Make an Interactive Model

What is data science and data science pipeline

From Data to Dashboard: Make an Interactive Model
2022年11月24日
00:00:00 - 00:11:00
Problem - From Data to Dashboard: Make an Interactive Model

Problem

From Data to Dashboard: Make an Interactive Model
2022年11月24日
00:11:00 - 00:20:00
Data - From Data to Dashboard: Make an Interactive Model

Data

From Data to Dashboard: Make an Interactive Model
2022年11月24日
00:20:00 - 00:35:02
Context - From Data to Dashboard: Make an Interactive Model

Context

From Data to Dashboard: Make an Interactive Model
2022年11月24日
00:35:02 - 00:48:05
Exploration - From Data to Dashboard: Make an Interactive Model

Exploration

From Data to Dashboard: Make an Interactive Model
2022年11月24日
00:48:05 - 00:55:10
Feature Engineering - From Data to Dashboard: Make an Interactive Model

Feature Engineering

From Data to Dashboard: Make an Interactive Model
2022年11月24日
00:55:10 - 01:02:40
Preprocessing - From Data to Dashboard: Make an Interactive Model

Preprocessing

From Data to Dashboard: Make an Interactive Model
2022年11月24日
01:02:40 - 01:08:12
Modeling - From Data to Dashboard: Make an Interactive Model

Modeling

From Data to Dashboard: Make an Interactive Model
2022年11月24日
01:08:12 - 01:14:07
Optimization - From Data to Dashboard: Make an Interactive Model

Optimization

From Data to Dashboard: Make an Interactive Model
2022年11月24日
01:14:07 - 01:15:53
Evaluation and Validation - From Data to Dashboard: Make an Interactive Model

Evaluation and Validation

From Data to Dashboard: Make an Interactive Model
2022年11月24日
01:15:53 - 01:20:38
Dashboarding - From Data to Dashboard: Make an Interactive Model

Dashboarding

From Data to Dashboard: Make an Interactive Model
2022年11月24日
01:20:38 - 01:23:17
Conclusion - From Data to Dashboard: Make an Interactive Model

Conclusion

From Data to Dashboard: Make an Interactive Model
2022年11月24日
01:23:17 - 01:27:38
Introduction - Topology for Time Series

Introduction

Topology for Time Series
2022年11月17日
00:00:00 - 00:00:42
Time Series Data - Topology for Time Series

Time Series Data

Topology for Time Series
2022年11月17日
00:00:42 - 00:05:09
Topology - Topology for Time Series

Topology

Topology for Time Series
2022年11月17日
00:05:09 - 00:07:13
Homology - Topology for Time Series

Homology

Topology for Time Series
2022年11月17日
00:07:13 - 00:09:00
Comparing Time Series with Persistent Homology - Topology for Time Series

Comparing Time Series with Persistent Homology

Topology for Time Series
2022年11月17日
00:09:00 - 00:10:55
Dataset - Topology for Time Series

Dataset

Topology for Time Series
2022年11月17日
00:10:55 - 00:16:40
Live R Coding - Topology for Time Series

Live R Coding

Topology for Time Series
2022年11月17日
00:16:40 - 00:25:30
QnA - Topology for Time Series

QnA

Topology for Time Series
2022年11月17日
00:25:30 - 00:35:58
Introduction - How to Validate your Model & Data and Easily Avoid Common ML Pitfalls

Introduction

How to Validate your Model & Data and Easily Avoid Common ML Pitfalls
2022年11月16日
00:00:00 - 00:03:15
ML Failures and Motivation - How to Validate your Model & Data and Easily Avoid Common ML Pitfalls

ML Failures and Motivation

How to Validate your Model & Data and Easily Avoid Common ML Pitfalls
2022年11月16日
00:03:15 - 00:18:10
ML Validation/Testing - How to Validate your Model & Data and Easily Avoid Common ML Pitfalls

ML Validation/Testing

How to Validate your Model & Data and Easily Avoid Common ML Pitfalls
2022年11月16日
00:18:10 - 00:24:35
Deepcheck Packages - How to Validate your Model & Data and Easily Avoid Common ML Pitfalls

Deepcheck Packages

How to Validate your Model & Data and Easily Avoid Common ML Pitfalls
2022年11月16日
00:24:35 - 00:29:15
Live Code Example - How to Validate your Model & Data and Easily Avoid Common ML Pitfalls

Live Code Example

How to Validate your Model & Data and Easily Avoid Common ML Pitfalls
2022年11月16日
00:29:15 - 00:46:32
QnA - How to Validate your Model & Data and Easily Avoid Common ML Pitfalls

QnA

How to Validate your Model & Data and Easily Avoid Common ML Pitfalls
2022年11月16日
00:46:32 - 00:53:24