Provides tools for working with nonlinear least squares problems. It is intended to eventually supersede the 'nls()' function in the R distribution. For example, 'nls()' specifically does NOT deal with small or zero residual problems as its Gauss-Newton method frequently stops with 'singular gradient' messages. 'nlsr' is based on the now-deprecated package 'nlmrt', and has refactored functions and R-language symbolic derivative features.
