Possible bug in cell.py

[macul99]

I guess the line #100 of cell.py should be 'z1' instead of 'z0':

"f = Activation(self.recurrent_activation)(z0)"

[rtygbwwwerr]

I think so, the code probably is wrong: i = Activation(self.recurrent_activation)(z0) f = Activation(self.recurrent_activation)(z0)

    c = add([multiply([f, c_tm1]), multiply([i, Activation(self.activation)(z2)])])
    o = Activation(self.recurrent_activation)(z3)
    h = multiply([o, Activation(self.activation)(c)])
    y = Activation(self.activation)(W2(h))

Apparently, in this code, f would equal to i, which contradicts with LSTM model.

[dsx4602]

I can't agree more. z1 should be instead of z0.