-
Notifications
You must be signed in to change notification settings - Fork 0
Open
Description
- Establish a new training loop architecture for your RNN/CNN approach.
- Aim to get training working with this new architecture, with val/train loss being calculated to have an idea of model accuracy.
- Once you have finished the base model, look into adding some simple features on top of that, like LR scheduling. Also feel free to experiment with different levels of data augmentation for your input.
Metadata
Metadata
Assignees
Labels
No labels