Open zmfb039 opened 6 years ago
Just realised that my default training was for the period of "day", thus all my _zoo files contain run-params for "day". This part makes a bit more sense now.
If you have the resources I would suggest you run and predict the price constantly. You can make the move when the model is most certain about the next price.
This is a fascinating project, nice work. I have managed to get everything to run but have a couple of questions? Do you have a recommended time for training the data (for the first time)? Secondly can you provide a bit more information about the best time to run the model, presumably at the start of the day to predict the remainder of the day? Following on from this is there any way to run the predict.py with something longer than the default day? I have tried changing this but cannot get anything to run.
Hope my questions make some sort of sense. Thanks for a great project and any additional info.