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 2.2.04Jul32130Jun2625242322191817141311103230May29282722212019161514652130Apr18161411109843131Mar2928272524232221201714121110765432128Feb272625232017151413121110631Jan2724232221161513109826Dec231916151411965432130Nov25242221209876542131Oct292726232218131143130Sep27212016429Aug279731Jul2522211917161513119854130Jun29262522211915141310729May221716151413107129Apr282524Post train quant: Save per-layer quantization params to yaml fileDump outputs in run log dir instead of script dirSummaryGraph: Changes in adjacency_map and predecessors/successorsBugfix in normalize_module_nameUpdated READMEget_dummy_input: extend to return tuples of tensors + add testsLogging: Capture warnings + optional verbose loggingADC: move the loggers to a separate filefeatures_collector.py: change the log level for the log encapsulating the stats collection phaseauto_compression_channels.yaml: Fix ResNet50 and add MobileNetv2cleanup the Coach interfaceAdded an example of faster Resnet20 training scheduleUpdate spinningup interfaceTD3 preset updateChange the CIFAR10 dataset normalization valuesAlways set a value for the seedPrivate: remove hard-coded valuesADC: only save checkpoints of generated networks if requestedAdded --amc-reconstruct-ptsCheckpoint loading: allow loading non-strict state-keys (#300)Fix citation links in README.mdAdded citations to the READMEactivation stats collection: add more layer types to default module listSimulated BN fold module changesSummaryGraph: Add adjacency map + numerous changes (#291)Simulated BN folding during training (module impl only) (#274)model_summaries: fix weights in compute-summaryMerge branch 'master' into new_amcprivate_if - cleanupspinningup fixesADC: too much to detailconfigure the dataset sampling configurationthinning: change logging level from info to debugGreedy filter pruning: add mobilenet_v1 greedy pruningCleanup the auto_compression_xxxx.yaml filessummary_graph: fix the calculation the number of incoming channelsAdd the value of PYTHONPATH to the logModel drawing: allow rendering models w/o specifying the dataset (#294)Input shape attribute in models (#292)update requirements.txt