Vessel_Trajectory_Prediction

Unofficial Implementation of Paper: Deep Learning Methods for Vessel Trajectory Prediction based on Recurrent Neural Networks

Important

To use this repo:

• Import necessary modules

  import torch
  from vTrack.vTrack import vTrack

• Declare hyper-parameters

  input_size = 4
  output_size = 2
  hidden_size = 16
  num_layers = 3
  seq_length = (12, 6) # 12 input sequences, 6 output sequences
  batch_size = 8

• Input sample

  x = torch.rand(batch_size, seq_length[0], input_size) # (8, 12, 4)

• Create object of the Model

  model = vTrack(input_size, hidden_size, num_layers, output_size, seq_length[1])

• Feed the sample to the Model

  out = model(x) # (batch, out_seq_length, output_size) : (8, 6, 2)

Warning

Errors and Omissions expected!