search

TimyadNyda / Variational-Lstm-Autoencoder

Lstm variational auto-encoder for time series anomaly detection and features extraction
MIT License
311 stars 61 forks source link

Visualization of latent space #11

Open Mechov opened 2 years ago

Mechov commented 2 years ago

Hi @TimyadNyda ,

How could we visualize the latent space using your code?

Thanks.

TimyadNyda commented 2 years ago

Hi @Mechov,

You could query the latent space with the function "reduce" of the encoder. Depending on the latent dimension you choose, you must use a dim reduction algorithm (t-sne, umap, pca, ...) over it to get a dim of 2 or 3 and plot it :)

Hope this helps !