Sign in to confirm you’re not a bot
This helps protect our community. Learn more
These chapters are auto-generated

Introduction

0:00

About KerasCV

1:20

Object Detection

4:10

Bounded Box Conversion

5:10

Numerical Optimization

9:00

Loading Data

9:35

Data Augmentation

11:30

Layer Mixup

12:55

Load Pretrained Models

15:00

Compile Model

15:50

Train Model

16:40

Custom Data Augmentation

17:30

Inference

20:10

Open Questions

21:47

Segmentation

22:42

Augmentation

25:10

Model initialization

27:00

YOLO model

27:55

Model fit

28:50

Modify backbone

29:17

Changing the backbone

30:43

Build a new model

30:55

Task presets

31:48

Visualization

34:51

Jax

37:22

Data Augmentation Example

41:42

How toContribute

43:15

Question

45:58

How to get into Open Source

54:36

What makes Open Source amazing

55:31

I love working in Open Source

57:25

Closing remarks

57:53
Introduction to KerasCV with Google Software Engineer | PyImageSearch | LiveStream
The podcast will form a significant part of the "Keras Community Days," an initiative to grow and nurture the vibrant open-source community around Keras. Here, we'll engage in meaningful discussions about open-source contributions and the best practices to embark on this rewarding journey. This is your chance to understand the open-source landscape better, interact with an expert, and even get your hands dirty with your first (or next) open-source contribution. The first guest of this podcast is Ian Stenbit. Ian is a software engineer at Google leading the KerasCV project, an effort to make training and serving modern computer vision models easy with Keras. Ian has worked at Google in Boulder, Colorado for 4 years. Prior to his time at Google, he completed a Masters in Computer Science + ML at Southern Methodist University. He is an avid ML practitioner and Keras user, and is a strong believer in open-source as the best way forward for the global ML community. Follow along the podcast on your own using this colab notebook: https://colab.research.google.com/gis... Directions for use: 1. Make a copy of the notebook 2. Run through with it alongside the video 3. Verify results #KerasDays #KerasCommunityDays #LiveSTream #LiveLearning #KerasCV #Keras #TensorFlow

Follow along using the transcript.

PyImageSearch

19K subscribers