Releases
v1.1.0
Added
Unit tests and removed system CUDA dependency;
Support for Conv1d and Conv3d Layers;
Legacy support;
MANIFEST.in and resolved header dependency;
Native toggle for forward_legacy and size arguments to tune;
codecov
integration;
Support for all torch.distributions
;
1R programming routine and non-linear device simulation during inference;
Stanford PKU and A Data Driven Verilog a ReRAM memristor models;
Modular crossbar tile support;
ADC and variable input voltage range support, and modularized all memtorch.mn
modules;
cibuildwheel
integration to automatically generate build wheels.
Enhanced
Mapping functionality;
Reduced pooling memory usage with maxtasksperchild
;
Programming routine;
set_conductance
;
apply_cycle_variability
.
Fixed
Dimension mismatch error for convolutional layers with non-zero padding;
reg.coef_
and reg.intercept_
extraction process for N-dimensional arrays;
Various semantic errors.
You can’t perform that action at this time.