2. Importing data¶

Tip

It'll be easier to keep the VM in fullscreen.

You can open the tutorial website inside the VM as you're working through it so you don't have to switch in and out of fullscreen mode.

To do this, open the applications menu from the Ubuntu icon in the top-left corner, launch Google Chrome (it's under Favorites), and navigate to tutorial.sleap.ai.

The input to SLEAP are video files, so we'll first need to load some data to work with.

For this tutorial, we've prepared a video that is representative of a typical behavioral setup in the lab. To access it, just double-click the sleap-tutorial-data folder in the desktop:

The mice.mp4 will be the main video we will use for training a model in this tutorial. The new_data folder will be used later in step 6.

Note

If you are working locally, you can get this data by cloning the data repository:

git clone https://github.com/talmolab/sleap-tutorial-data

You do not need to run this in the VM!

Import videos into SLEAP¶

Open a terminal by right-clicking the background of the tutorial data folder and selecting Open Terminal Here:

Then launch the SLEAP GUI by typing this command followed by Enter:
```
sleap
```
Once loaded, the SLEAP GUI should appear like this:
Go to File → Add Videos... to open the file browser.
Select mice.mp4 and click Open.
The video import interface will appear. Click Import to finish adding the videos.

You'll now see the first video in the GUI:

Configure skeleton¶

Go to the Skeleton tab on the right (it's already selected when SLEAP opens), then click on New Node three times to create three entries in the table.
For each node, double-click on the name to edit it, then Enter to save.

Name them head, torso, and tail_base.
Switch to the Edges tab and add connections: torso → head, and torso → tail_base.

The final skeleton setup should look like this:

You did it!

Next up: Initial labeling