1. Learn
    2. /
  2. Courses
    3. /
  3. Intermediate Deep Learning with PyTorch

Exercise

Building a forecasting RNN

It's time to build your first recurrent network! It will be a sequence-to-vector model consisting of an RNN layer with two layers and a hidden_size of 32. After the RNN layer, a simple linear layer will map the outputs to a single value to be predicted.

The following imports have already be done for you:

import torch
import torch.nn as nn

Instructions 1/4

undefined XP
    1
    2
    3
    4
  • Define the RNN layer passing it the correct values for input_size, hidden_size, num_layers, and batch_first, and assign it to self.rnn.