flexclustControl-class       package:flexclust       R Documentation

_C_l_a_s_s_e_s "_f_l_e_x_c_l_u_s_t_C_o_n_t_r_o_l" _a_n_d "_c_c_l_u_s_t_C_o_n_t_r_o_l"

_D_e_s_c_r_i_p_t_i_o_n:

     Hyperparameters for cluster algorithms.

_O_b_j_e_c_t_s _f_r_o_m _t_h_e _C_l_a_s_s:

     Objects can be created by calls of the form
     'new("flexclustControl", ...)'. In addition, named lists can be
     coerced to 'flexclustControl' objects, names are completed if
     unique (see examples).

_S_l_o_t_s:

     Objects of class 'flexclustControl' have the following slots:

     '_i_t_e_r._m_a_x': Maximum number of iterations.

     '_t_o_l_e_r_a_n_c_e': The algorithm is stopped when the (relative) change
          of the optimization criterium is smaller than 'tolerance'.

     '_v_e_r_b_o_s_e': If a positive integer, then progress is reported every
          'verbose' iterations. If 0, no output is generated during
          model fitting.

     '_c_l_a_s_s_i_f_y': Character string, one of '"auto"', '"weighted"',
          '"hard"' or '"simann"'.

     '_g_a_m_m_a': Gamma value for weighted hard competitve learning.

     '_s_i_m_a_n_n': Parameters for simulated annealing optimization (only
          used when 'classify="simann"').

     '_n_t_r_y': Number of trials per iteration for QT clustering.

     '_m_i_n._s_i_z_e': Clusters smaller than this value are treated as
          outliers.

     Objects of class 'cclustControl' inherit from 'flexclustControl'
     and have the following additional slots:

     '_m_e_t_h_o_d': Learning rate for hard competitive learning, one of
          '"polynomial"' or '"exponential"'.

     '_p_o_l._r_a_t_e': Positive number for polynomial learning rate of form
          1/iter^par.

     '_e_x_p._r_a_t_e' {Vector of length 2 with parameters for exponential
          learning rate of form Vector of length 2 with parameters for
          exponential learning rate of form
          eqnnormal-bracket62bracket-normal.

     '_n_g._r_a_t_e': Vector of length 4 with parameters for neural gas, see
          details below.

_L_e_a_r_n_i_n_g _R_a_t_e _o_f _N_e_u_r_a_l _G_a_s:

     The neural gas algorithm uses updates of form

                 cnew = cold + e*exp(-m/l)*(x - cold)

     for every centroid, where m is the order (minus 1) of the centroid
     with respect to distance to data point x (0=closest, 1=second,
     ...). The parameters e and l are given by

                e = par1*(par2/par1)^(iter/iter.max),


                l = par3*(par4/par3)^(iter/iter.max).

     See Martinetz et al (1993) for details of the algorithm, and the
     examples section on how to obtain default values.

_A_u_t_h_o_r(_s):

     Friedrich Leisch

_R_e_f_e_r_e_n_c_e_s:

     Martinetz T., Berkovich S., and Schulten K (1993). `Neural-Gas'
     Network for Vector Quantization and its Application to Time-Series
     Prediction. IEEE Transactions on Neural Networks, 4 (4), pp.
     558-569.

_S_e_e _A_l_s_o:

     'kcca', 'cclust'

_E_x_a_m_p_l_e_s:

     ## have a look at the defaults
     new("flexclustControl")

     ## corce a list
     mycont = list(iter=500, tol=0.001, class="w")
     as(mycont, "flexclustControl")

     ## some additional slots
     as(mycont, "cclustControl")

     ## default values for ng.rate
     new("cclustControl")@ng.rate

