Kalyan Veeramachaneni
Kalyan Veeramachaneni
Opening a thread to discuss ```NASA``` dataset. We can add comments and discuss its characteristics, suitability for using it for benchmarking.
Starting a thread to discuss ```Yahoo``` dataset. We can add our comments, discuss its characteristics, and its suitability for being used as a dataset to benchmark unsupervised time series anomaly...
We can add Weibull distribution for univariate. It is adaptive and can capture multiple shapes. https://en.m.wikipedia.org/wiki/Weibull_distribution I was looking for it’s implementation in scipy stats.
The ```bayesmark``` package is another ```wrapper``` hyper parameter tuning library. We can add this to our benchmarking suite. Per their documentation, they wrap around: ``` The builtin optimizers are wrappers...
A question about your TGAN code. We need to avoid any randomness for reproducibility purposes. We were wondering if there is a way to avoid this? We have tried to...
Add pywavelets as primitive. Seems to be a well maintained library for wavelets. https://github.com/PyWavelets/pywt worth examining @csala
Surprise provides a variety of solutions for recommender systems. It has SVD, SVD++, NMF, Slope One, k-NN, Centered k-NN, k-NN Baseline, Co-Clustering, and Random. It is pretty close to scikit-learn...