Skip to content
Snippets Groups Projects

Repository graph

You can move around the graph by using the arrow keys.
Select Git revision
  • 4-bit-qat
  • api_changes
  • ax_integration
  • gste
  • inception_support
  • kd_fixes
  • lattice_regularization
  • load_quant_aware_train_checkpoint
  • master default protected
  • ptq_propagate_settings
  • scale_approximation
  • update_links
  • v0.3.2
  • amc_before_pytorch1.1_merge
  • v0.3.1
  • v0.3.0
  • v0.2.1
  • v0.2.0
  • v0.1.0
19 results
Created with Raphaël 2.2.015May14652130Apr18161411109843131Mar2928272524232221201714121110765432128Feb272625232017151413121110631Jan2724232221161513109826Dec231916151411965432130Nov25242221209876542131Oct292726232218131143130Sep27212016429Aug279731Jul2522211917161513119854130Jun29262522211915141310729May221716151413107129Apr282524Merge branch 'master' of https://github.com/NervanaSystems/distillerSummaryGraph: fix ‘weights_vol’ attribute for conv and linear layersActivation Histograms (#254)SummaryGraph changes: _force_outplace + OrderedDictsFix failing test after last post-train quant changeImproved logging when saving collectors data (#251)Post train quant - warn instead of exit when dynamic quant not supportedMerge branch 'master' of https://github.com/NervanaSystems/distillerFix broken load test (#245)AMC: Revive support for weights-channels removalAllow loading checkpoints not containing an optimizerQuantizer: Proper handling of modules that point to same object (#239)Update READMEPost-train quant: Ensure quant params are located on correct device (#241)Added PackedSequence functionality (#236)Remove single worker limitation in deterministic mode (#227)LSTM: Modular implementation + Post-Train Quantization Sample (#196)ActivationStatsCollectors: Handle non-contiguous tensors (#228)Post-train quant: Extend acts clipping functionality (#225)Docs: Fix broken images and linkssave_checkpoint(): Save dict of user values (Breaking) (#219)Add functionality to log values of buffers in a model (#220)SummaryGraph - fix MACs calculation for grouped-convolutionsRelaxation of the SummaryGraph API (#212)Documentation: add missing imagesRefine pruning logic (#222)Proper handling of 0s in DoReFa 1-bit weights (#205)Removed sys.path modifications when importing distiller. (#224)Fix issue #213 (#221)Add FP16 support in post-training quantization (#216)lsq for weightsgstegsteBugfix in create_model()Update and organize .gitignorelsq activationsQuantizer: Specify # bias bits + custom overrides (BREAKING) (#178)setup.py - correct distiller versionsetup.py - fix recursive importChange distiller version to "0.4.0-pre"Add ResNeXt101 pruning example (#201)Load optimizer from checkpoint (BREAKING - see details) (#182)