This is a very simple problem, I'm looking for someone with experience with LSTM or any other algorithm (like it could be XGB / Random Forest) in order to improve the performance of a deep learning network.
Dataset: 520,000 records.
Attached to this project, you'll find the sample & training data + the script which is already created.
The main goal, is to know if the value will to know if the next value will be bigger than 20, or lower. Even the chances are good.
(1) You'll need to modify it to read from CSV instead, replacing "read_sql_query" for "read_csv"
(2) I've been trying only with one column, which is "value", and that's (included "[login to view URL]")
(3) I've also included a 2ND CSV file including "id" + "value" + "creacion" (included in "[login to view URL]")
Feel free to ask for any questions! Anyone who can increase the rate, specially, predict when the value will be lower than 20 is already good enough!
20 freelancers are bidding on average $196 for this job
hey, I hope you are having a base script for the same. kindly share the same so that I can actually work on what has to be changed to increase the accuracy of your prediction in lstm.