Use back propogation to adjust the weight matrices w and v to make the predictions more accurate.
Use back propogation to adjust the weight matrices w and v to make the predictions more accurate. The implementation uses vector operations.
http://www4.rgu.ac.uk/files/chapter3%20-%20bp.pdf
Return the fit (weigth matrix w, weigth matrix v, bias vector wb, bias vector vb)
Show the flaw by printing the error message.
Show the flaw by printing the error message.
the method where the error occurred
the error message
Given several input vectors z_i, predict the output/response vector z_o(0).
Given an input vector z_i, predict the output/response scalar z_o(0).
Given a new discrete data vector z, predict the y-value of f(z).
Given a new discrete data vector z, predict the y-value of f(z).
the vector to use for prediction
Given several input vectors z_i, predict the output/response vectors z_o.
Given several input vectors z_i, predict the output/response vectors z_o.
the new input vectors (stored as rows in a matrix)
Given an input vector z_i, predict the output/response vector z_o.
Given an input vector z_i, predict the output/response vector z_o.
the new input vector
Set the initial weight matrices w and v randomly with a value in (0, 1) before training.
Set the initial weight matrices w and v randomly with a value in (0, 1) before training.
the random number stream to use
Set the initial weight matrices w and v manually before training.
Set the initial weight matrices w and v manually before training.
the initial weights for w
the initial weights for v
the initial bias for wb
the initial bias for vb
Return the vector of values of the sigmoid function applied to vector t.
Return the vector of values of the sigmoid function applied to vector t.
the sigmoid function vector argument
Return the value of the sigmoid function at t.
Return the value of the sigmoid function at t.
the sigmoid function argument
Given training data x and y, fit the weight matrices w and v.
This class supports basic 3-layer (input, hidden and output) Neural Networks. Given several input and output vectors (training data), fit the weights connecting the layers, so that for a new input vector z_i, the net can predict the output vector z_o, i.e., z_i --> z_h = f (w * z_i) --> z_o = g (v * z_h)