Zineb Adaika
Reputation: 1
I faced an error when I used PCA with LSTM model
I have a time series dataset with 20 classes, but they are imbalanced; when I tried a method like "RandomOverSampler", I got an error because of the 3D of our data so could you suggest a method that can work with 3D data or what I should do in this case? Also, how can I use PCA as a feature extraction method with LSTM? Here, I also faced the same issue of dimensionality; the output of PCA is 2D, but LSTM needs 3D as an input. What should I do? However, will the accuracy of using PCA with LSTM improve?
Upvotes: 0
Views: 34
Answers (0)
Related Questions
- LSTM time series data classification model low accuracy
- Dimension of data before and after performing PCA
- PCA for Recurrent Neural Networks (LSTM) - Shall I use PCA for target variables too?
- Optimal Feature Selection Technique after PCA?
- When should we use Principal Component Analysis?
- Performance drop using PCA with LSTM
- Keras LSTM model data reshapping