achevrot
commented
2 years ago The windows have a fixed length in the whole dataset I used, regardless of the context they were taken from. As a result, there are more windows from the cruising phase than from the others. But this is not a problem as you can just take a fixed number of windows for each phase during training, or use a sample weight.
Then you set the shape of the input tensor of the CAE according to the window size you chose for your dataset.
As mentioned in Section 4.1, some external contextual factors may alter the shape of the time windows over time, hence every time window is associated with a contextual feature. I think this may mean that the three different flight phases have different window sizes. However, in Fig.2 (architecture of CAE model), the window size is fixed. Could you explain in detail how to set the window size in the CAE model?