August 7
-
The metric is calculated and all the error value are obtained from the data.
- One more possibility can be there, what if there are more than variables that are dependent on each other.
-
In this case the we have to change our optimizer and shift it to do a cartesian product of all the variable.
-
we are restricting ourself with the Gradient descent as it is giving as a much better result than the earlier, with all the optimised value of the variables.
- the GUI is developed in a modular style, so it is easy in the future to implement different optimiser with little change in the code.
Task for tomorrow:
- Analysis on the data generated.