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 2.2.025Jun24232221191817151412111098543228May272625242221201918151413121187654130Apr2928Merge remote-tracking branch 'upstream/dev' into feature/Pool_Batch_opMerge remote-tracking branch 'upstream/dev' into feature/check_int_input_before_infer_ConvInpGenMerge remote-tracking branch 'upstream/dev' into feature/add_downsampler_opMerge remote-tracking branch 'upstream/dev' into feature/fix_globalaccpool_batch_shapeMerge remote-tracking branch 'upstream/dev' into feature/Add_channelwise_op[Test] Add rtlsim[FPGADataflow] Add scalar parameter broadcast to InferChannelwiseLinearLayer[Test] Update test for MoveFlattenPastAffine to use data layout[Streamline] Add condition that data layout is set to NHWC for MoveFlattenPastAffine trafo[Test] Add test for MoveFlattenPastAffine[Streamline] Add MoveFlattenPastAffine operation[Test] Add test for AbsorbTransposeIntoFlatten trafo[Streamline] Add AbsorbTransposeIntoFlatten trafo[Transform] skip and not fail when const params not foundMerge pull request #165 from Xilinx/feature/QuantAvgPool2dNHWC[Streamline] in MoveMulPastDWConv raise warning if mul weights are not set[Test] Extend change_datalayout test to check layout annotation after InferDataLayout transform[CustomOp] Update QuantAvgPool2d[Transform] Add QuantAvgPool to infer datalayout trafo[Streamline] Update MoveMulPastDWConv transformationMerge branch 'dev' into feature/move_chw_mul_past_dw_conv[Test] Add test for mv transpose node past scal mul[Streamline] Add trafo to move transpose node past scalar mul[FPGADataflow] Add check to InferConvInpGen to avoid inference is input is not integer[CustomOp] Update make_shape_compatible fct in QuantAvgPool2d to guarantee that InferShapes works directly after brevitas export[Transform] Update change datalayout trafo for quantavgpool[CustomOp] Integrate NHWC functionality into QUantAvgPool2d node[Registry] Remove obsolete QuantAvgPool2dNHWC nodeDelete obsolete files[Test] Add test for quantavgpool change datalayout trafo[TEST] Add idt== DataType.INT32 test case[FPGADataFlow] Add to skip data type list DataType.FLOAT32. Set maximum width supported data type instead and issue warningMerge branch 'dev' into feature/QuantAvgPool2dNHWC[FPGADataflow] Fix typo -> Add DataType.TERNARY to skip list[Test] Add test for InferChannelwiseLinearLayer[FPGADataflow] Fix InferChannelwiseLinearLayer. Correct output datatype inference from idt and parameters. Add parameter type inference. Check parameter shape. Check if retrived intializer is None (dynamic input)[HLSCustomOp] Add paramDataType attribute for the user to control the parameter datatype and be able to visualize it once the op is infered[HLSCustomOp] remove unroll limits in hls synthesis tcl[StreamingFC] get_weightstream_width is 0 except for decoupledMerge commit 'f00c5c14a36e4422698e63e2565ad7ba754b68a6' into feature/Add_channelwise_op