A simple and flexible code for Reservoir Computing architectures like Echo State Networks
numpy.memmap
concurrent accesses. Which fixes many issues with parallelization. The memmap name is now unique for each buffer.
Fixes #141, fixes #112, fixes #57mat_gen
can now take a degree
argument instead of connectivity
.Delay
node, which returns its input with a specified delay.ScikitLearnNode
introduced in ReservoirPy v0.3.10. This has breaking changes, as the previous implementation was not consistent with the library node API. The tutorial has subsequently been updated.dataset.narma
now takes a u
parameter so that the input timeseries can be accessed. Fixes #142ValueError
. Fixes #138Node.partial_backward
method of offline nodes didn't concatenate output timeseries correctly.Concat
node(by @Deepayan137 )
ESN.run
had the same problem as ESN.fit
when using multiprocessing
as joblib backend.Full Changelog: https://github.com/reservoirpy/reservoirpy/compare/v0.3.9...v0.3.9.post1
japanese_vowels
link was dead, the dataset could not be loaded (#113).multiprocessing
as joblib backend was failing because of a locale function in ESN.fit
.dataset
module coherence and precision: Add **kwargs
everywhere, fix some issues with timeseries length (#118 #117)Full Changelog: https://github.com/reservoirpy/reservoirpy/compare/v0.3.8...v0.3.9
lr
in reservoir nodes can now be an array or list of float values, one per neurons.Full Changelog: https://github.com/reservoirpy/reservoirpy/compare/v0.3.7...v0.3.8
h
parameter can now be changed in datasets.mackey_glass
.noise_kwargs
parameters to reservoir nodes (Reservoir and IPReservoir) to change noise distribution parameters.noiseFull Changelog: https://github.com/reservoirpy/reservoirpy/compare/v0.3.6...v0.3.7
datasets
!FORCE
class is now split in two: RLS
and LMS
classes, to avoid confusion between FORCE algorithm and RLS/LMS learning rules. FORCE
is still available but deprecated.Full Changelog: https://github.com/reservoirpy/reservoirpy/compare/v0.3.2...v0.3.4
japanese_vowels
to datasets
! You can now dowload the Japanese vowels dataset to try Reservoir Computing on a well known pattern recognition task.ArpackNoConvergence
Exception (#67)Concat
nodes were not created at the right place) (#68)Model.run
can now be called on multiple series (using lists of arrays or arrays with ndim > 2)Lock
was used too soon in Ridge.partial_fit
. Dot product is now performed in parallel.Unsupervised
node subclass can be used to mark a Node as unsupervised learner.numpy.random.Generator
in hyper.research
.Full Changelog: https://github.com/reservoirpy/reservoirpy/compare/v0.3.2...v0.3.3