nemos.callbacks.SolverConvergenceCallback#

class nemos.callbacks.SolverConvergenceCallback[source]#

Delegate convergence checking to the solver’s built-in criterion.

Calls ctx.solver.stochastic_convergence_criterion(...) at the end of each pass and requests a stop if it returns True.

Tracks previous params and state internally so the context doesn’t have to.

Methods

`__init__`()
`on_batch_begin`(ctx)	Run before each batch update.
`on_batch_end`(ctx)	Run after each batch update.
`on_pass_begin`(ctx)	Save current params and state before the pass runs.
`on_pass_end`(ctx)	Check solver convergence criterion and request stop if met.
`on_train_begin`(ctx)	Run once at the start of training.
`on_train_end`(ctx)	Run once at the end of training.

on_batch_begin(ctx)#

Run before each batch update.

on_batch_end(ctx)#

Run after each batch update.

on_pass_begin(ctx)[source]#

Save current params and state before the pass runs.

on_pass_end(ctx)[source]#

Check solver convergence criterion and request stop if met.

on_train_begin(ctx)#

Run once at the start of training.

on_train_end(ctx)#

Run once at the end of training.