dsipts.models.Persistent module

class dsipts.models.Persistent.Persistent(future_steps, past_steps, loss_type=None, persistence_weight=0.1, optim_config=None, scheduler_config=None, **kwargs)[source]

Bases: Base

Persistent model propagatinng last observed values

Parameters:

  • future_steps (int) – number of future lag to predict

  • past_steps (int) – number of future lag to predict. Useless but needed for the other stuff

  • optim_config (dict, optional) – configuration for Adam optimizer. Defaults to None. Usless for this model

  • scheduler_config (dict, optional) – configuration for stepLR scheduler. Defaults to None. Usless for this model

handle_multivariate = True
handle_future_covariates = False
handle_categorical_variables = False
handle_quantile_loss = False
description = 'Can   handle multivariate output \nCan NOT  handle future covariates\nCan NOT  handle categorical covariates\nCan NOT  handle Quantile loss function'
__init__(future_steps, past_steps, loss_type=None, persistence_weight=0.1, optim_config=None, scheduler_config=None, **kwargs)[source]

Persistent model propagatinng last observed values

Parameters:

  • future_steps (int) – number of future lag to predict

  • past_steps (int) – number of future lag to predict. Useless but needed for the other stuff

  • optim_config (dict, optional) – configuration for Adam optimizer. Defaults to None. Usless for this model

  • scheduler_config (dict, optional) – configuration for stepLR scheduler. Defaults to None. Usless for this model

forward(batch)[source]

It is mandatory to implement this method

Parameters:

batch (dict) – batch of the dataloader

Returns:

result

Return type:

torch.tensor