lefnire / tforce_btc_trader

TensorForce Bitcoin Trading Bot
http://ocdevel.com/podcasts/machine-learning/26
GNU Affero General Public License v3.0
820 stars 235 forks source link

Use conv2d depth (channel / z-axis) instead of height (y-axis) for features #2

Closed lefnire closed 6 years ago

lefnire commented 6 years ago

Currently the conv2d set uses time as the x-axis (naturally), and features as the y-axis (not-so-naturally). Then we have window/stride sort of finding patterns positionally. Makes perfect sense to have a window finding positional patterns on the time axis, but really the features should all boil down together, not in chunks. So TODO: experiment with features as channels/depth instead of height.