Search Results for "training"

...A time series dataset class that abstracts handling variable transformations, missing values, randomized subsampling, multiple history lengths, etc. A base model class that provides basic training of time series models along with logging in tensorboard and generic visualizations such actual vs predictions and dependency plots. Multiple neural network architectures for timeseries forecasting that have been enhanced for real-world deployment and come with in-built interpretation capabilities. The package is built on PyTorch Lightning to allow training on CPUs, single and multiple GPUs out-of-the-box.

...Country holidays & recurring special events. Sparsity of coefficients through regularization. Plotting for forecast components, model coefficients as well as final predictions. Automatic selection of training-related hyperparameters.