akshayka
commented
3 years ago Sure, you can embed that with PyMDE.
Is the idea that you want to embed each sample? In that case, you can convert your dataframe into a NumPy array of shape (50, 1000), and use pymde.preserve_neighbors or pymde.preserve_distances to obtain an embedding for each sample.
The documentation explains how to do things like this. Let me know if you have follow-up questions, or if something in the documentation is not clear.
for example if I have a dataframe shape of (50,1000), 50 samples with time series length of 1000 each