From the course: Apache Spark Deep Learning Essential Training
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Transfer learning in action - Spark DataFrames Tutorial
From the course: Apache Spark Deep Learning Essential Training
Transfer learning in action
- [Instructor] So, we're going to do transfer learning using the ResNet50 model. Now, the ResNet50 model has been trained on the ImageNet data set, which has 1,000 classes. So, we've got a couple of varieties of flowers in the flower data set that we'll use. We've got daisies, dandelions, roses, sunflowers, and tulips. Now, it so happens that daisies are one of the 1,000 categories in ImageNet, so we won't pick them. Instead we'll use transfer learning to distinguish between two classes of flowers, tulips and sunflowers. So, let's head over to the notebook, so I select Workspace, I select the down arrow, I select Import, Browse, I select the notebook 04_03_begin. I select Import, so we'll do our usual checks, confirm that we've got the correct libraries installed. I select Attach and Run to attach this notebook to the cluster. And then what you're going to do is to download the flower data set. Enter and unzip it, go into the flower photos and just confirm that we've got the different…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.