Timothy Yu

EDIT: However, if the `sigma` and `uthreshold` values don't match when reversing the denoising process, **then the reconstructed signal is going to be invalid** - and since there are clearly...

@JannyKul Your comment(s)/criticism of the model design and existing attempts to replicate/implement said model are valid - they do not need retraction. For a streaming/online model, a dynamic `sigma` &...

incomplete/work in progress: https://github.com/timothyyu/wsae-lstm/commit/9c796bf4e46558e5e7e5526dfec9763a286b74a4#diff-3c3d6d5243e1476c8c1f21078c759772R39

this is directly related to #5 a dual-stage approach is probably required due to the nature of the dataset (complex hierarchical time series with numerical features), so a simple transformation...

Copied from #5 (dataset scaling/normalization before wavelet transform): The author of [`DeepLearning_Financial`](https://github.com/mlpanda/DeepLearning_Financial/blob/master/run_training.py) decided to forgo automated scaling/normalization and instead scaled the input features/dataset manually before applying the wavelet transform: https://github.com/mlpanda/DeepLearning_Financial/blob/7e846144629d8b49b8fd74a87d5ff047b7af55d1/run_training.py#L55...

RobustScaler test for `'nifty 50 index data'`:     

see commit https://github.com/timothyyu/wsae-lstm/commit/8073c426f903611f7ec22043d4b5378054b2904b scaled data and scaled denoised data now saved in data/interim folder:  pdf output of train-validate-test split scaled + denoised in reports folder:  Excerpt from pdf...

train-validate-test split is showing some questionable output; look into when I get a chance there is the possibility it's a matplotlib/pdf render output issue: https://github.com/timothyyu/wsae-lstm/blob/master/reports/djia%20index%20data%20tvt%20split%20scale%20denoise%20visual.pdf this should not be happening:...

upon closer examination, it appears that the line/feature flatling visually is not technically wrong for the `djia index` dataset:     The values are still there, but `RobustScaler`...

@mg64ve the `subrepos` directory contains `deeplearning_financial` for reference; `wsae_lstm` is the main source location for the code for my implementation. I am using the following directory structure, but with `wsae_lstm`...