Results from 18-hours of training a machine learning model | Airbnb Amenity Detection 7

Опубликовано: 13 Январь 2025
на канале: Daniel Bourke

15,096

222

I trained a machine learning model for 18-hours on ~35,000 images on a single GPU (P100) using transfer learning. These are the results.

Although we didn't reach Airbnb's requirement for mean average precision (metric for measuring object detection models), the project still goes on.

Next will be using the full model to build a small application around so others can use it and get it deployed.

Stay tuned for more.

All videos in this series:
Full playlist - https://dbourke.link/airbnbplaylist
Part 1 (overview) -    • Using Machine Learning to Replicate A...
Part 2 (data exploration) -    • Using Machine Learning to Replicate A...
Part 3 (data manipulation) -    • Replicating Airbnb's Machine Learning...
Part 4 (building a small dog model) -    • I got Detectron2 working! | Airbnb Ma...
Part 5 (experiment tracking) -    • Every model I build has to answer 1 s...
Part 6 (modelling experimentation) -    • 34,835 training images downloaded | R...
Part 7 (building a big dog model) -    • Results from 18-hours of training a m...
Part 8 (getting my modelling deployed) -    • I got my machine learning model deplo...
Part 9 (project wrap up and retrospective) -    • What could be improved? Machine Learn...

Note: I am not affiliated with Airbnb nor are they involved with this series of videos in any way.

Links:
My machine learning course - https://dbourke.link/mlcourse
All of my notes/ideas in Notion - https://dbourke.link/airbnb42days
Airbnb Amenity Detection article -   / 05qperz3a4

Get email updates on my work - https://dbourke.link/newsletter
Support on Patreon - https://bit.ly/mrdbourkepatreon

Connect elsewhere:
Web - https://dbourke.link/web
Quora - https://dbourke.link/quora
Medium - https://dbourke.link/medium
Twitter - https://dbourke.link/twitter
LinkedIn - https://dbourke.link/linkedin

#machinelearning #datascience #deeplearning