Skip to content
Snippets Groups Projects
Select Git revision
  • DevPTQ
  • DevQAT
  • Fix_ptq
  • MulCompensation_Nodes
  • channel_wise_quant
  • dev
  • fix_ci
  • leaky_relu_support
  • main default protected
  • mobilenetv1_support
  • qat
  • quantization_cast
  • real_quantization_cast
  • refact_removing_scaling_nodes
  • scaling_node_for_rescale_tensor
  • storing_sf_as_producer_attributes
  • tag_instead_of_metaop
  • tag_true_quantization
  • test_LeNet
  • true_quantization
  • v0.1.2
  • v0.1.1
  • v0.1.0
23 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.025Jun24231917116318May86530Apr292822181716151110432131Mar2827252019181716141211654328Feb27242120191714115131Jan3029282723222119171513863220Dec1817161310953229Nov28272625181331Oct30292823211698743227Sep26231896543230Aug23229865231Jul302926252423221918161283128Jun2019121110727May23222114136229Apr24178528Mar824Jan191822Nov151424Oct236425Sep2221145429Aug2423222111854Adding assert to check dataformatchannel_wise_qu…channel_wise_quantMerge branch 'dev' of https://gitlab.eclipse.org/eclipse/aidge/aidge_quantization into dev re-removing insertcompensationnode() from hpp fileTag int8/ adding the cast routine and testing it with minirestnettag_true_quanti…tag_true_quantizationMerged with devFix unit testNew nodes inherit data formatContinuing work on Tag (Finshed CastQuantizedNetwork RoutineAdding auto tag function + repalced nbBits parameterRemoving checksum from test_ptq (keeping it local)Apply 1 suggestion(s) to 1 file(s)Fix requested change 1/2updating .gitigoredeleting unwanted filesAdding properly handled insertCompensationNodes()Activating cuda in setup.pydeleting unwanted filesUnits testsContinuin Unit TestIncluding new architectures for test pipelineunits testFix unit testFixing bad files modoficationfix_cifix_ciNew nodes inherit data formatMerge branch 'rounding_modes' of gitlab.eclipse.org:eclipse/aidge/aidge_quantization into rounding_modesuse the new rounding systemhandle halfway rounding modes + scaling factor setteruse the new rounding systemUse OperatorTensorWithImplVersion 0.4.0 -> 0.4.1showQuantizationVersion now use fmt print instead of Log::info.Fix documentation link.HotFix: Segfault in the PTQ when using a model without biasWithout a CUDA compiler pip install . produces the following error:Merge branch 'main' into devRemove CIBW test as they are currently stuck.Update pyproject.toml with search path to debug locally.Update build dependencies.udpate cibuildwheel_build_deps_before_build_wheel.sh.Update manylinux image used by CIBW to manylinux 2.28.
Loading