MLP model

[martins0n]

🚀 Feature Request

Add MLPModel model for etna deep model family

Proposal

• You can use as ref

class MLPNet(DeepBaseNet):
    def __init__(
        self,
        input_size: int,
        hidden_size: List[int],
        lr: float,
        loss: "torch.nn.Module",
        optimizer_params: Optional[dict],
    ) -> None:
   pass

• For MLP we dont need encoder part: so encoder_lenght=0
• Typical batch sample would have shape [encoder_lenght, input_size]

Test cases

• add unit tests for make_sample
• add seasonal overfit test
• add test for step and retrun values
• add test for forward and return values

Additional context

No response