Timothy Yu

@mg64ve I am looking into this exact issue in implementing the WSAE-LSTM model, which uses the wavelet transform to denoise data (Bao et al., 2017): https://github.com/timothyyu/wsae-lstm My implementation is a...

Here's an example of applying the wavelet transform to the first two `train-validate-test` splits of the `csci300 index data` : 

@mg64ve yes, this is from my own code. I have an updated implementation of the above (scaling is done on the train set, and then applied to the validate and...

@mg64ve here is an updated version of the above that clearly illustrates the `train-validate-test` split, with the effect of scaling and scaling + denoising being visualized: > Implemented as of...

Scaling is done with `RobustScaler` on the `train` set, and then the same parameters used to scale the `train` set are applied to the `validate` and `test` sets. ddi_scaled[index_name][intervals from...

incorporates https://github.com/dysonance/Trendy/issues/1 (already merged)

potential to include as view/output as part of #18

related to https://github.com/timothyyu/gdax-orderbook-ml/issues/12

maya package for dealing with epoch/datetime/tz: https://github.com/kennethreitz/maya

potential application of lstmvis project to model layer weights + cell states: http://lstm.seas.harvard.edu/ https://github.com/HendrikStrobelt/LSTMVis Related: https://github.com/keras-team/keras/issues/4962 https://github.com/cbaziotis/neat-vision Attention mechanism (partially related): https://github.com/philipperemy/keras-attention-mechanism