DeepDream is a computer vision program created by Alexander Mordvintsev which uses a CNN to find and enhance patterns in images. It uses the Caffe neural network framework trained on the ImageNet dataset to produce "dream visuals".
Image taken from deepdreamgenerator
We will use it today to turn a picture of dinosaurs into flowers (inspired by Chris Rodley's project: Turning Dinosaurs into Flowers)
1. Create a Workspace
Workspaces are created using the
Create Workspace at the top right of the Workspaces section of the web console.
Name your workspace and initialize the code from
Select machine type
K80 and the
We don't need to add any datasets, so just click
Confirm and Start:
Confirm that your workspace details and then
3. Run the Jupyter Notebook!
Congratulations! - your notebook is now running on Spell! Navigate to deepdream folder and then open the
Follow the steps in
dream.ipynb to get some lovely looking dinosaurs:
Feel free to experiment with your own images by uploading them to the Notebook!