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 2.2.013Mar12111096543228Feb2726252423212019181716141312111076543131Jan30292827242322212017161514109873220Dec19181716131211109654[Transform] add DoubleToSingleFloat transformation[Test] Add mem mode as pytest variable for rtlsimMerge branch 'dev' into feature/weight_streamers_rtlsim[StreamingFC] Generation of .dat file into generate_params function[finn-rtllib] Renamed file ramb18.v to ramb18_wf_dualport.v[Util] add other PYNQ boards to pynq_part_map[Streamline] add Absorb1BitMulIntoConv into Streamline[StreamingFC] Save weights array into .dat file and save in verilog folderFix image link in blog postRename 2020_03_11-rn50-released.md to 2020-03-11-rn50-released.mdMerge pull request #59 from quetric/masterBlog post for rn50 releaseAdded illustration for RN50 IPI-based compilation[StreamingFC] Copy rtllib memstream verilog files automatically after ip block is generated for MVAU in decoupled mode[Streamline] add a first v of Absorb1BitMulIntoConv (untested)[Streamline] better 1d checking for absorbption[Test] add a streamlining test for cnv-w1a1[Streamingfc] Add options for rtlsim decoupled mode in corresponding streamingfclayer functions[Streamingfc & Test] Commit after clean-up by pre-commit[Test] Add test cases for decoupled mode[Streamingfclayer] Edit template parameters for code generation decoupled modeMerge branch 'dev' into feature/weight_streamersMerge branch 'feature/update_finn_hlslib' into dev[StreamingFC] change TDstI generation for new finn-hlslib[Docker] pull and use latest v of finn-hlslib masterMerge branch 'feature/fix_conv_bn2affine' into dev[Test] update conv bn2affine test with actual execution[BN2Affine] handle BN param shapes better[Test] bring back conv for batchnorm2affine testMerge branch 'dev' into feature/weight_streamers[Source code] Make copy of array before saved to .npy fileMerge branch 'feature/conv_lowering' into dev[Test] compare against expected output in conv lowering test[CnvLower] add qnt annotations to newly created tensors[CnvLower] fix output tensor name in shape inference[Test] Add test for mem_mode "decoupled"[StreamingFC] Make copy of weights before saving them to .npy file[Streaming FC & test] added experimental dimension fliiping[Test] start adding a test for conv lowering[CnvLower] access int array attributes correctly