FuSAGNet Running Analysis

1. Original Train Dataset

1.1 Shape of original train dataset

$$\left[\begin{array}{ccccccccc}0.935538& 0.433560& 1.0& 1.0& \dots & 0.000037& 0.0& 0.0& 0.0\\ 0.894236& 0.434363& 1.0& 1.0& \dots & 0.000037& 0.0& 0.0& 0.0\\ 0.952981& 0.433002& 1.0& 1.0& \dots & 0.000037& 0.0& 0.0& 0.0\\ 0.912204& 0.434542& 1.0& 1.0& \dots & 0.000037& 0.0& 0.0& 0.0\\ 0.922004& 0.433872& 1.0& 1.0& \dots & 0.000037& 0.0& 0.0& 0.0\\ ⋮& ⋮& ⋮& ⋮& \ddots & ⋮& ⋮& ⋮& ⋮\\ 0.958814& 0.457039& 1.0& 1.0& \dots & 0.000073& 0.0& 0.0& 0.0\\ 0.919379& 0.457976& 1.0& 1.0& \dots & 0.000073& 0.0& 0.0& 0.0\\ 0.911795& 0.456012& 1.0& 1.0& \dots & 0.000073& 0.0& 0.0& 0.0\\ 0.954614& 0.455633& 1.0& 1.0& \dots & 0.000073& 0.0& 0.0& 0.0\\ 0.898320& 0.457418& 1.0& 1.0& \dots & 0.000073& 0.0& 0.0& 0.0\end{array}\right]$$

• The shape is [47520, 51]

2. TimeDataset

• The final return

class TimeDataset(Dataset):
	pass
	def __getitem__(self, idx):
		pass
		return window, window_y, label, edge_index

• window: [51, 5]
• window_y: [51]
• label: [51]
• edge_index: [2, 2550]

2.1 Shape of train TimeDataset

• This is train_dataset which used to generate train_dataloader
• In train_dataloader author use random to split train subset and valid subset
• Therefore, we will get different train and valid dataset in every signal training

$$[\begin{array}{c}[\begin{array}{ccc}{\left[\begin{array}{c}9.5726e-01\\ 4.5503e-02\\ ⋮\\ 5.8423e-01\end{array}\right]}^t& {\left[\begin{array}{c}9.1500e-01\\ 4.7642e-02\\ ⋮\\ 2.0000e-01\end{array}\right]}^{t+1}& [\end{array}\end{array}$$