IBM/aihwkit

Issues

• Per-Column Weight Scaling

  #688 opened 2 months ago by coreylammie
  1

• Model Initialized outsize [w_min, w_max]: Pinpointing the bug in issue #604

  #635 opened 7 months ago by Zhaoxian-Wu
  7

• AnalogConv2d incorrectly converts the input size

  #641 opened 2 months ago by Zhaoxian-Wu
  5

• Gradient error on GPU for subsequent Conv2d utilization after downsampling.

  #662 opened 5 months ago by KyleM-Irreversible
  4

• Backward pass not ideal when using InferenceRPUConfig

  #628 opened 2 months ago by HeatPhoenix
  4

• Version issues during pip installation in the Advanced Installation Guide

  #689 opened 2 months ago by sheep1222
  3

• Underlying simulation

  #671 opened 3 months ago by chaseltb
  1

• How to Configure ReRAM as an RPU

  #684 opened 3 months ago by 24367452
  2

• How to Configure ReRAM as a Processing Unit

  #683 opened 3 months ago by 24367452
  2

• Analog Context doesn't get casted to float16 when model.to(dtype=float16) is called

  #680 opened 3 months ago by jubueche
  1

• drift on non-backpropagation based algorithms

  #634 opened 8 months ago by imanehmz
  9

• Sample Validation Failure

  #669 opened 3 months ago by 24367452
  6

• Improve documentation and add an example of training with half precision

  #677 opened 3 months ago by PabloCarmona
  0

• Create new build wheels for osx

  #676 opened 3 months ago by PabloCarmona
  0

• Improve installation docs

  #675 opened 3 months ago by PabloCarmona
  0

• Segment Fault Happens when Applying Gradient Update on TransferCompound on CUDA in Debug Mode

  #632 opened 8 months ago by Zhaoxian-Wu
  2

• Code modification and application method

  #620 opened 3 months ago by DoyoonKim0209
  5

• For convolutional layers, `convert_to_analog` and `convert_to_analog_mapped` behave differently

  #627 opened 8 months ago by coreylammie
  2

• CMake building problems

  #670 opened 4 months ago by amine-ziad-ounnoughene
  2

• AIHWKIT installation failed

  #668 opened 4 months ago by 24367452
  0

• AIHWKIT installation error

  #667 opened 4 months ago by 24367452
  0

• Aihwkit installation failed

  #666 opened 4 months ago by 24367452
  0

• Training with half-precision doesn't work for the torch tile or CUDA bindings

  #623 opened 5 months ago by coreylammie
  10

• Continuing training based on checkpoint using torch tile

  #609 opened 5 months ago by jubueche
  7

• AnalogConv2d fails when using TT-v2

  #642 opened 5 months ago by Zhaoxian-Wu
  5

• undefined read noise in PCM inference model

  #660 opened 6 months ago by matifali
  1

• WeightModifierType.DROP_CONNECT not implemented

  #622 opened 8 months ago by VictorY-1Qbit
  4

• Erroneous reference to "ADD_NORMAL" in WeightNoiseType

  #608 opened 8 months ago by HeatPhoenix
  2

• Considering weight updates for half-biased devices

  #619 opened 8 months ago by Seungkun-Kim
  4

• `WeightModifierType.DISCRETIZE` does not work on CUDA

  #614 opened 8 months ago by jubueche
  17

• fusion_import function modifies the input model with the new conductance data

  #613 opened 8 months ago by kkvtran
  3

• Question about IOParameters arguments

  #616 opened 9 months ago by VictorY-1Qbit
  0

• Analog conversion for Large models

  #593 opened 10 months ago by WickedStereo
  7

• ImportError of libmkl_intel_lp64.so.2 during compiling

  #597 opened 10 months ago by Zhaoxian-Wu
  9

• Model Initialized outsize [w_min, w_max]

  #599 opened 10 months ago by Zhaoxian-Wu
  7

• Documentation referring to "aihwkit.simulator.devices" which does not exist

  #592 opened 10 months ago by HeatPhoenix
  2

• "Misplaced &" in the document

  #596 opened 10 months ago by Zhaoxian-Wu
  2

• Got runtime error when loading state_dict for AnalogLinearMapped layer

  #595 opened 10 months ago by kkvtran
  2

• Error occurred while building dockerfile

  #566 opened 10 months ago by JongchanRyu
  1

• GPU conda installation is not working properly

  #584 opened 10 months ago by jubueche
  5

• Cannot use analog models on multiple GPUs with torch.nnDataParallel

  #578 opened a year ago by jdbuhler
  7

• Error when compiling aihwkit on mac: fail to find pybind11

  #545 opened a year ago by Zhaoxian-Wu
  13

• Transfer compound as optimizors

  #560 opened a year ago by Zhaoxian-Wu
  3

• Question about aihwkit install

  #574 opened a year ago by YuqiL1
  2

• Discrepancy Between Documentation and Code: AnalogModuleBase

  #559 opened a year ago by Zhaoxian-Wu
  3

• Consider regenerating the api stubs

  #577 opened a year ago by diego-plan9
  0

• CUDA enabled IBM kit Advanced development setup

  #579 opened a year ago by adnanrana88
  2

• Using analog models in DDP with multiple GPUs causes warnings and errors related to "wrong device"

  #565 opened a year ago by jdbuhler
  1

• cannot use the cuda and gpu after compile

  #549 opened a year ago by meicale
  3

• Update to torch 2?

  #550 opened a year ago by meicale
  1