Gompertz curve optimization
Equation of the Gompertz model:
With the solution:
For physical usage, we add time shift parameter:
We define a set of training points
For given parameters:
Residues of the differencial equation:
which leads to the following error:
And the classic error of the data, MSE of between the simulated data and the training ones:
Finaly the loss is:
Then, the selection of
We use
and add some noise, a Gaussian distribution of standard deviation of 5% of