AutoDL is a repository that automates the process of deep learning model selection, architecture search, and hyperparameter tuning using a Bayesian optimizer. It aims to find the best deep learning model for a given task by efficiently exploring the search space of different architectures and hyperparameters.
AutoDL utilizes a Bayesian optimizer, such as Optuna or Hyperopt, to efficiently search the hyperparameter space and architecture space of deep learning models.
The repository provides a predefined search space for various deep learning models, including architectures like Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Transformer models.
AutoDL performs hyperparameter tuning for each architecture using the Bayesian optimizer. It automatically searches for the optimal hyperparameters, such as learning rate, batch size, activation functions, optimizer, etc.
The models are trained with early stopping to prevent overfitting and to save time during the optimization process.