News for hackers. Stuff that matters.
Re:Stock Price Prediction with LSTMs by bcheung on Wednesday March 21, 2018 @05:05PM (#16642408)
Cool to see that you created a notebook and are publishing this like a paper. I was thinking of doing some projects and this looks like a good format to follow.

However, this looks to be grossly overfitting. You can't just randomly drop out samples in a time series and use those for the test dataset. You need to cut out larger contiguous sequential time ranges and reserve those for your test set. Probably a single contiguous time window.

Anyone can predict with a high degree of accuracy what stock price is given the last 5 days and the next 5 days.

Predicting a few randomly dropped out pixels in that graph is really easy due to the nature of time series. You can just interpolate it.

Re:Stock Price Prediction with LSTMs by highd on Wednesday March 21, 2018 @05:07PM (#16642421)
It's actually using just one day's data (one point) to predict the next day's data, let alone predicting 5 days.