In addition to univariate signals, we would like to include multivariate signals as possible inputs to the pipeline. The general format of the data will follow the following structure
timestamp dim 0 dim 1 dim 2 1222819200 -0.366358 0 0 1222840800 -0.394107 1 0 1222862400 0.403624 1 1 1222884000 -0.362759 0 0 1222905600 -0.370746 1 0The usage of the Orion class will stay the same, we need to only alter the variable input_shape to allow multiple channel ingestion. For now, we assume that we are only predicting / reconstructing a single channel (MISO).
from orion import Orion
hyperparameters = {
'orion.primitives.tadgan.TadGAN#1': {
'input_shape': [100, 3], # where 100 is the window size and 3 is the number of channels
'verbose': True
}
}
orion = Orion(
pipeline='tadgan',
hyperparameters=hyperparameters
)
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4