I often put my phone on DND and only allow calls , that really helps. From morning 9 to 6 and everytime I am on my bed. Removing social media apps also helps. I wont say its 100% effective but I see drastic reduction in my screen time.
I have true hate for the authors of this paper: "A deep learning framework for financial time series using stacked autoencoders and long-short term memory". Probably the most complex AND vague in terms of methodology and after weeks trying to reproduce their results (and failing) I figured out that they were leaking future data into their training set (this also happens more than you'd think)."
- Not sure how author tried to implement it , but is this not how you train LSTM networks by feeding t+1 data back into the cell again to predict t+2 data. It will be easier if author made it open source as well
I wrote my thesis last year comparing different RNNs against each other using this exact paper as baseline and basically concluded that you would be better off predicting the price yesterday than using their results. Authors did not respond when prompted for implementation details or comments.
Overall, concluded that amongst RNNs the GRU architecture proved most favorable but still would not outperform simple stochastic models of the financial industry toolbox.
Leaking future data in would be using t+1 for t, e.g. something like a bi-directional LSTM. I assume he means the actual training dataset had some kind of signal in the data that was also in the test data.
People do this by doing things like testing that their features contain information in both the training and test set. Because they are not exposing the data directly to the classifier they think that they haven't compromised the test set - but what they have done is increased the chances of a chance correlation.
This will impact the tourism industry to some extent and also high time for governments across the world to take positive steps in the same direction as this will eventually affect all of us. I stay in India and recent news related to water scarcity across cities are trending which is not usual.
I was really looking forward to probabilistic programming but then I saw this project is based on Julia. Not to keen on leaning Julia as I already know Python/C++/Java/Scala/R. Wish it was based on c++/python to at least get started with. Please let me know if there are other interesting probabilistic programming frameworks available or tutorials to get started with.
Julia is fast becoming one of the great programming languages of our era. Julia is not going away, its usage is only becoming more pervasive. Imagine if you had learned Python 20 years ago. Julia's founding sponsor is Dr. Jeremy Kepner [1] who is the founder of MIT Lincoln Laboratory Supercomputing Center [2] and the founder of GraphBLAS [3]. Kepner's got a good eye for things, he knows what's coming down the pipe and what will be important in a language of the future. Learning Julia now means you'll be ahead of the curve for years to come. Your time will not be wasted.
