Project
A project defines the scope of the models and model versions for your use case.
Last updated
A project defines the scope of the models and model versions for your use case.
Last updated
You can create a project after logging on to your account.
To create a project:
Click the side menu on the top left.
Click New Project.
On the New Project window, enter a name for your project.
Enter a description for the project.
Click Save.
To open and start working on your project, click the side menu on the top left again, click Open Project, then select the project by name.
When you save your project for the first time, Tensorleap treats it as the master branch, with subsequent revisions being new branches. You can create different branches when saving a new version of a project. For more information, see Versions.
The project layout has the Versions, Network and Dashboard views. Also on the top bar are the Train and Evaluate buttons, where you set the Training and Evaluation Plans for a model (see Evaluate/Train Model).
The Project Layout has three views:
The video below shows how to navigate the Views.
Versions - Allows saving your network into different versions (see Versions). Accessed by clicking .
Network - Where you lay out the layers and other component blocks, including the data block, visualizer, loss and optimizer blocks, in your network (see Network). Accessed by clicking .
Dashboard - Where you can add and view the analysis being performed on a model, and observe and query various metrics (see Analysis). Accessed by clicking .