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 2.2.010Jun98543228May272625242221201918151413121187654130Apr2928272624232221201917161514987Merge branch 'feature/stream_perf_analysis' into dev[Test] add vcd perf checks into test_end2end_tfc_w1a1[Util] add vcd utils for FIFO count signals[Util] extra comment re. fifo cnt trace depth[FIFO] expose FIFO count port[Util] approx clock edges for if stats[rtlsim] no autoeval for IP-stitched rtlsim[Deps] update pyverilator[Core] bugfix in execute_node[Docker] Update brevitas commit version[Test] Add test for AbsorbConsecutiveTransposes[Transformation] add AbsorbConsecutiveTransposes to remove consecutive opposite transposes[Test] Add signed input and not completely folded input test cases[HLSCustomOp] fix input folding != complete folding. Remove restriction on signed DataTypes as is the problem is fixed in hlslib. Add Variable output width feature as hlslib now supports it[Test] higher precision for isclose comparision of produced and expected output[CustomOp] Change make_shape_compatible fct of QuantAvgPoolMerge dev into feature/onnx_export_quant_avg_pool[Test] Update avg pool test[CustomOp] Add node execution functionality to QuantAvgPool2d[Util] parallelize get_all_stream_if_stats[Util] introduce get_all_stream_if_stats[Util] add comment about trace depths[Util] bugfix in list_stream_if, use vars for stream signal names[Util] add vcd.py with some stream performance analysis util fxns[Reqs] add vcdvcd to requirements.txt[FPGADataflow] Add InferGlobalAccPoolLayer transform to support the implementation of GlobalAveragePool[HLSCustomOp] Fix computed output shape of GlobalAccPool_Batch[Test] Add test for ChannelwiseOp_Batch op (required to change hlslib commit)[FPGADataflow] Add InferChannelwiseLinearLayer Transform to get ChannelwiseOp_Batch nodes[HLSCustomOp] Add ChannelwiseOp_Batch for standalone functions with one parameter per channelMerge branch 'feature/conv_input_gen_dws' into dev[Docker] typo fix in run-docker.sh[CustomOp] add some comments to Im2Col and its HLS variantMerge branch 'dev' into feature/conv_input_gen_dwsMerge pull request #141 from quetric/feature/add_fmpadding_node[HLSCustomOp] remove OutputDim as FMPadding attribute, can compute[Test] convert_to_hls_conv_layer needs vivado, mark as such[Deps] update hlslib to get FMPadding[Test] Update QuantAvgPool test to compare results[CustomOp] Change calculation for execution slightly in QuantAvgPool2d