solvers page upd

docs/make.jl

@@ -16,7 +16,8 @@ makedocs(
             "basics/distributed.md",
             "table-formats/scenario.md",
             "table-formats/measurement.md",
-            "table-formats/parameters.md"
+            "table-formats/parameters.md",
+            "basics/solvers.md"
         ],
         "Tutorial" => [ # methods
             "tutorial/intro.md",

docs/src/basics/solvers.md

@@ -0,0 +1,20 @@
+# Solver choice
+
+### Page under development. Information will be updated
+
+## ODE solver choice
+
+`HetaSimulator.jl` relies on [DifferentialEquations.jl](https://docs.sciml.ai/DiffEqDocs/stable/) packages, which provide access to 300+ ODE solvers. 
+These solvers can be used in `HetaSimulator` simulation and parameters estimation functions (`sim`, `fit`, etc.), provided via `alg` keyword argument (For example, see [`sim`](@ref)). 
+One can also set relative and absolute tolerances (`reltol`, `abstol`) and other stepsize related settings via relevant keyword arguments. See [DiffEqDocs](https://docs.sciml.ai/DiffEqDocs/stable/basics/common_solver_opts/) for details.
+
+If no solver is provided the default one `AutoTsit5(Rosenbrock23())` will be used with `reltol=1e-3` and `abstol=1e-6` for simulation problems and `reltol=1e-6` and `abstol=1e-8` for parameters estimation.
+
+The following [DiffEq Docs page](https://docs.sciml.ai/DiffEqDocs/stable/solvers/ode_solve/) provides general advice on how to choose a solver suitable for your ODE system. 
+For most of ODE systems the following simplified guideline is sufficient:
+1. If your system is small (~10 ODEs) go with the default solver 
+2. If the system is large and stiff choose `FBDF()` or `QNDF()`
+3. If `FBDF()` fails to solve the system check the model and try it again :)
+4. If `FBDF()` still fails to solve the system or the integration takes too long, try `CVODE_BDF()`*
+
+* You should be caution with `CVODE_BDF()`. With many models it is the most fast solver, however the accuracy of the solution is often the tradeoff. You shouldn't use it with tolerances higher than ~ `reltol=1e-4` and `abstol=1e-7`.