Skip to content
Snippets Groups Projects
Select Git revision
  • accl default
  • adegendt/accl_clone
  • dev
  • full_accl_support
  • georg/accl-build
  • georg/accl-complete
  • georg/accl-finn
  • georg/end2end
  • main protected
9 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.04Jun3228May272625242221201918151413121187654130Apr2928272624232221201917161514987632131Mar3027[HLSCustomOp] Add first draft of vvau node[Docker] use correct hlslib url for finn_ci[Test] Add test for FMPadding_Batch HLSCustomOp[HLSCustomOp] Add HLSCustomOp for FMPadding_Batch HLS function call to perform padding[Test] Add test for LowerConvsToMatMul applied to conv with kernel size 1x1, stride == 1 and no padding[Transform] if kernel size is 1x1, stride == 1 and no padding is used, LowerConvsToMatMul doesn't add im2col[Test] Add test case for sliding window generator with kernel_size % stride != 0[HLSCustomOp] Add support for sliding window generator with kernel_size % stride != 0[Test] Add depthwise convolution cppsim/rtlsim test[HLSCustomOp] Add depthwise option to ConvInputGen custom opMerge pull request #130 from quetric/feature/SortGraphMerge pull request #135 from Xilinx/feature/sparsity_annotationMerge pull request #137 from Xilinx/feature/update_to_latest_finn-hlslib[HLSCustomOp] bring back directives option in base[HLSCustomOp] bring back directives option in hls_syn.tcl[StreamingFC] Update saving of weights for lates finn-hlslib streaming mvau[Transform] update comments for SortGraph[HLSCustomOp] Update ConvInputGen with default resource typeAdd toposort to requirements[CustomOp] Changing setting of resource type for ConvInputGen[Docker] finn-hlslib dependency repo to latest commit version[Transformation] Ignore "dw-layers" during hls conversion[Test] Extend test lowering of dw conv[Docker] Update brevitas commit version[Test] add rtlsim throughput test in end2end_tfc_w1a1[Core] introduce throughput_test_rtlsim[rtlsim] allow using larger batch inputs[Test] Check if information about dw conv is set in nodes after lowering trafo[CustomOp] Add new node attribute "dw" to Im2Col[Transformation] Add information about depthwise conv to created nodes when lowering convs to matmulMerge branch 'feature/sparsity_annotation' into feature/depthwise_convolution[Core - Modelwrapper] Change name of variable in sparsity utility fct[Core & Test] Add modelwrapper functions to get and set sparsity annotation[Transformation] Change creation of sparse matrix for lowering dw conv to matmulMerge branch 'dev' into feature/depthwise_convolution[Docker] Update brevitas commit versionMerge branch 'dev' into feature/onnx_export_channelwise_scaling[Test] Add tests for dw conv lowering[Transformation] First draft of lowering depthwise convs to matmulMerge remote-tracking branch 'upstream/dev' into feature/MoveMaxPoolPastMultiThreshold