Modifications/Cleanup/Improvements to Time-Stepped Circuit IRDrop example and model implementation#753
Merged
PabloCarmona merged 13 commits intoIBM:rc-1.1.0from Jan 27, 2026
Merged
Conversation
* fix: debug mode, chopper trans bug, memory issue, maximizer trans * fix memeory error with valgrind * changelog * fix UMH test * CUDA arch
* update notebooks wheel to 1.0.0 gpu enabled Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update analog fusion notebook to use cuda Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add ipywidgets top lenet5 notebook Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update and run notebooks Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com>
Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com>
…IBM#744) Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com>
* modify CMakeLists to link properly extensions_ops and cuda together Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove strict flag to solve Py reference errors Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com>
charlesmackin
requested review from
PabloCarmona,
atvasilopoulos,
charlesmackin and
Copilot
Contributor
There was a problem hiding this comment.
Pull request overview
This PR enhances the time-stepped IR Drop model implementation with new features including ADC quantization and input-dependent PCM read noise modeling, along with comprehensive example testing and a license migration from MIT to Apache 2.0.
Key Changes
- Added ADC quantization feature with configurable frequency parameter to model Current-Controlled Oscillator (CCO) behavior
- Implemented input-dependent PCM read noise model with analytical fit coefficients based on integration time and conductance values
- Enhanced example 28 to test new IR Drop model parameters including PCM noise and ADC quantization flags
- Migrated license headers from MIT to Apache 2.0 across modified files
- Refactored
_matmul_irdropmethod to use trapezoidal integration instead of summation by default
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 13 comments.
| File | Description |
|---|---|
| src/aihwkit/simulator/tiles/analog_mvm_irdrop_t.py | Core implementation of ADC quantization and PCM read noise models; refactored integration logic; updated license header |
| src/aihwkit/simulator/parameters/io.py | Added new parameters for xdep PCM read noise, ADC quantization, and integration method selection; updated license header and simplified documentation |
| examples/28_advanced_irdrop.py | Extended example to test new features with multiple model configurations; updated license header |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
jluquin
force-pushed
the
jluq/irdrop
branch
2 times, most recently
from
731231f to
5e766b3
Compare
Collaborator
|
@jluquin, can you please address the left comments by Copilot if necessary and change the target branch to the |
Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com>
…on; signoff included Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com>
Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com>
Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc715.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com>
Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com>
PabloCarmona
added a commit
that referenced
this pull request
Jan 27, 2026
…mple and model implementation (#753) * Fix chopper bug (#732) * fix: debug mode, chopper trans bug, memory issue, maximizer trans * fix memeory error with valgrind * changelog * fix UMH test * CUDA arch * Add files via upload (#739) * update notebooks wheel to 1.0.0 gpu enabled (#741) * update notebooks wheel to 1.0.0 gpu enabled Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update analog fusion notebook to use cuda Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add ipywidgets top lenet5 notebook Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update and run notebooks Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * change online demo link to proper one (#743) Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> * remove release-build.yaml to change to new and stablish build process (#744) Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> * Extensions fix (#745) * modify CMakeLists to link properly extensions_ops and cuda together Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove strict flag to solve Py reference errors Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * Modifications/Cleanup/Improvements to IRDrop example and implementation Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * Modifications/Cleanup/Improvements to IRDrop example and implementation; signoff included Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * PR modifications and corrections Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * PR modifications/update Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc715.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * Update src/aihwkit/simulator/parameters/io.py Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * Update src/aihwkit/simulator/parameters/io.py Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * PR modifications/update; re: minor copilot suggestions Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc715.pok.ibm.com> Co-authored-by: Malte Rasch <158572058+maljoras-sony@users.noreply.github.com> Co-authored-by: Corey Lammie <coreylammie@gmail.com> Co-authored-by: pablocarmona <pablocarmonagonzalez@gmail.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc710.pok.ibm.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc715.pok.ibm.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com>
PabloCarmona
added a commit
that referenced
this pull request
Feb 3, 2026
* Fix chopper bug (#732) * fix: debug mode, chopper trans bug, memory issue, maximizer trans * fix memeory error with valgrind * changelog * fix UMH test * CUDA arch Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * Add files via upload (#739) Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * update notebooks wheel to 1.0.0 gpu enabled (#741) * update notebooks wheel to 1.0.0 gpu enabled Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update analog fusion notebook to use cuda Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add ipywidgets top lenet5 notebook Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update and run notebooks Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * change online demo link to proper one (#743) Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * initial rc for v1.1.0 Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * remove release-build.yaml to change to new and stablish build process (#744) Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * Extensions fix (#745) * modify CMakeLists to link properly extensions_ops and cuda together Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove strict flag to solve Py reference errors Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * fix matplotlib install for Python 3.10 Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * fix matplotlib install for Python 3.10 Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * [fix] for #627 Signed-off-by: Julian Buechel <jub@zurich.ibm.com> Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * [deprecation] deprecate convert_to_analog_mapped Signed-off-by: Julian Buechel <jub@zurich.ibm.com> Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * changelog for convert_to_analog_mapped Signed-off-by: Julian Buechel <jub@zurich.ibm.com> Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * Create build-wheels.yml (#738) * Create build-wheels.yml Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * Update build-wheels.yml Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * Update build-wheels.yml Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * Update build-wheels.yml Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * Update build-wheels.yml Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * Update build-wheels.yml Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * Update build-wheels.yml Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add build for cuda Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove blas install Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * change torch version Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * change torch version Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove torch version for cpu and add openblas-devel for gpu Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update before all to match image repo Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * fix before all command Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * install only blas and gcc Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove sudo Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update before all to setup cuda env Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * change to one line Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * modify and add more setup and clean up for optimized space Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove exports Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add version specifier and clean up rpm file Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove cuda workflow and fixed latest torch version for cpu on tag push Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * Extensions fix (#745) * modify CMakeLists to link properly extensions_ops and cuda together Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove strict flag to solve Py reference errors Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add matrix of python versions Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * fix CI command Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * fix CIBW_BUILD env Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add cibw matrix Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * fix only python 3.10 3.11 and 3.12 Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * removed fixed version for pytorch Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove matrix Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove matrix Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add cleanup on before_build Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * split versions on jobs Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * change name of jobs Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * fix job names for build and update test and lint Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update cibw linux image Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add trigger to test build Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * change name for wheels artifact Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * revert to tag push Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Co-authored-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * fix pycodestyle errors on python 3.12 Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * remove python 3.12 Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * remove python 3.12 from lint Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * add cuda build workflow job Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * add trigger for cuda Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * remove cuda workflow Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * add 3.12 to test and lint Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * update pylint Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * update project toml and requirements deps versions Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * Added GlobalDriftCompensationWithExactReference drift compensation class (#674) * Add GlobalDriftCompensationWithExactReference class * Update CHANGELOG.md --------- Signed-off-by: Corey Liam Lammie <Corey.Lammie@ibm.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * add temp ignore for type checking Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * remove pylint errors with refactor and temp disabling in some methods Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * fix(pyproject.toml): move mypy to build-system req Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * change install to editable mode in gh actions test and lint Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * move mypy from build to deps Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * add temp flags for pylint to only look for errors Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * fix(quantized_base_modules.py): change import from nn to nn.functional to match proper layers Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * fix(pyproject.toml): bump version to 1.1.0 Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * add temp pylint disable not-callable Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * fix circular import for InferenceTileWithPeriphery while running examples Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * Run distilGPT2 on aihwkit (#754) * Fix chopper bug (#732) * fix: debug mode, chopper trans bug, memory issue, maximizer trans * fix memeory error with valgrind * changelog * fix UMH test * CUDA arch * Add files via upload (#739) * update notebooks wheel to 1.0.0 gpu enabled (#741) * update notebooks wheel to 1.0.0 gpu enabled Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update analog fusion notebook to use cuda Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add ipywidgets top lenet5 notebook Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update and run notebooks Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * change online demo link to proper one (#743) Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> * remove release-build.yaml to change to new and stablish build process (#744) Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> * Extensions fix (#745) * modify CMakeLists to link properly extensions_ops and cuda together Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove strict flag to solve Py reference errors Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add example 36 to implement distilgpt2 on the wikitext-2-raw-v1 dataset using AIHWKit Signed-off-by: anchen25 <an.chen@ieee.org> * change rpu_config to RPU_CONFIG in lines 340,342, 344 to correct check error Signed-off-by: anchen25 <an.chen@ieee.org> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> Signed-off-by: anchen25 <an.chen@ieee.org> Co-authored-by: Malte Rasch <158572058+maljoras-sony@users.noreply.github.com> Co-authored-by: Corey Lammie <coreylammie@gmail.com> Co-authored-by: pablocarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * update CHANGELOG with new example 36 Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * add tensorboard to examples deps and remove old requirements.txt setup Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * Modifications/Cleanup/Improvements to Time-Stepped Circuit IRDrop example and model implementation (#753) * Fix chopper bug (#732) * fix: debug mode, chopper trans bug, memory issue, maximizer trans * fix memeory error with valgrind * changelog * fix UMH test * CUDA arch * Add files via upload (#739) * update notebooks wheel to 1.0.0 gpu enabled (#741) * update notebooks wheel to 1.0.0 gpu enabled Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update analog fusion notebook to use cuda Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * add ipywidgets top lenet5 notebook Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * update and run notebooks Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * change online demo link to proper one (#743) Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> * remove release-build.yaml to change to new and stablish build process (#744) Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> * Extensions fix (#745) * modify CMakeLists to link properly extensions_ops and cuda together Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * remove strict flag to solve Py reference errors Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> * Modifications/Cleanup/Improvements to IRDrop example and implementation Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * Modifications/Cleanup/Improvements to IRDrop example and implementation; signoff included Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * PR modifications and corrections Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * PR modifications/update Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc715.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * Update src/aihwkit/simulator/parameters/io.py Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * Update src/aihwkit/simulator/parameters/io.py Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> * PR modifications/update; re: minor copilot suggestions Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> --------- Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc715.pok.ibm.com> Co-authored-by: Malte Rasch <158572058+maljoras-sony@users.noreply.github.com> Co-authored-by: Corey Lammie <coreylammie@gmail.com> Co-authored-by: pablocarmona <pablocarmonagonzalez@gmail.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc710.pok.ibm.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc715.pok.ibm.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * update CHANGELOG with improvements on ir_drop experiments Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * address lint errors with some ignores Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> * Fix: Added missing fields to bindings_ingore in IOParametersIRDropT, to fix failing test (test_storing_and_loading for TorchInferenceIRDropT (#756) Signed-off-by: jluquin@ibm.com;5J0442897;Jose Luquin <jluquin@ccc-login2.pok.ibm.com> Co-authored-by: jluquin@ibm.com;5J0442897;Jose Luquin <jluquin@ccc-login2.pok.ibm.com> --------- Signed-off-by: Pablo Carmona Gonzalez <pablocarmonagonzalez@gmail.com> Signed-off-by: PabloCarmona <pablocarmonagonzalez@gmail.com> Signed-off-by: Pablo <pablocarmonagonzalez@gmail.com> Signed-off-by: Julian Buechel <jub@zurich.ibm.com> Signed-off-by: Corey Liam Lammie <Corey.Lammie@ibm.com> Signed-off-by: anchen25 <an.chen@ieee.org> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Signed-off-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc715.pok.ibm.com> Signed-off-by: jluquin@ibm.com;5J0442897;Jose Luquin <jluquin@ccc-login2.pok.ibm.com> Co-authored-by: Malte Rasch <158572058+maljoras-sony@users.noreply.github.com> Co-authored-by: Corey Lammie <coreylammie@gmail.com> Co-authored-by: Julian Buechel <jub@zurich.ibm.com> Co-authored-by: anchen25 <an.chen@ieee.org> Co-authored-by: jluquin <40866771+jluquin@users.noreply.github.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc710.pok.ibm.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@ccc-login2.pok.ibm.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc585.pok.ibm.com> Co-authored-by: Jose Luquin jluquin@ibm.com <jluquin@cccxc715.pok.ibm.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Updated / Cleaned time-step IR Drop model functionality; including: ADC quantization feature (and corresponding io parameters) and new input-dependent PCM read noise model (and corresponding io parameters). Further, example 28 was extended to include more comprehensive (though simple) testing of IR Drop model flags/parameters.