Compute next gradient¶
A function in the MSData object to find the top signals in the MS1.
- computeNextGradient(self, acqFunc='ei')¶
Function to calculate the next gradient to run using an acquisition function.
- Parameters:
acqFunc – str. Acquisition functions in lower case. Valid options are: ei (expected improvement), explore (pure exploration), exploit (pure exploitation), eps (epsilon-greedy algorithm), pi (probability of improvement), rand (random a gradient), ucb (upper confidence bound)
To use this function:
# You need a gpModel object (model)
model.computeNextGradient()