Skip to content
Snippets Groups Projects
Select Git revision
  • clockwork default
  • clockwork-v0.6
  • master
  • v0.6.0
  • v0.5
  • v0.4
  • v0.3
  • v0.2
  • v0.1rc
9 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.06Oct5432129Sep2827262524232221201817131211108765432131Aug30292827262524232221201918171514131110987532131Jul28272625242320191817161514131211109876543130Jun292726232220181716151211105432131May30252423222120181716151412119876543229Apr272625242221201817161514121098231Mar2926251714131211753228Feb27262524222120171613121110987654231Jan302824232220191817161513121110865431Dec301112542129Nov28272625191817642128Oct272623222119181615141312[TOPI] add argmax, argmin (#515)Conv2d modified for better performance (#516)conv2d data re-layout fix out of threads bug (#514)[RUNTIME] Better error message in cuda launch (#513)conv2d adjusted to fix different workloads (#511)[SCHEDULE] Further fix of reduce inline with multiple outputs (#508)[SCHEDULE] Fix inline with multiple outputs (#507)[TOPI] Add left_shift, right_shift, clip, cast (#504)[LANG] Support for Bitwise Operation (#502)[LINT] Fix pylint (#501)[TOPI] Update depthwise conv2d schedule on rasp (#500)[DOCS] How to deploy TVM Modules (#499)fix squeeze to output (1,) if all axes are squeezed. E.g squeeze((1,1,1...), None) case (#498)[TOPI] Update conv schedule on rasp (#497)[TOP] Initial Schedule of MobileNet on Rasp (#496)[UTILS] Move target to tvm; rename convolution as conv2d (#492)[TOPI] add squeeze (#494)[RUNTIME] Fix graph runtime for gpu (#491)[CUDA] auto detect compatibility when arch is not passed (#490)fixed algorithm. padding should be on both sides (#489)[TOPI] add reshape, concatenate, split (#481)[ARITH] More robust int set checking (#487)[RUNTIME] Remove parameter def from runtime (#486)[DOC][BUILD] Fix cmake and docs (#485)[RUNTIME] Minimum graph runtime (#484)log_softmax added to topi (#483)fix fuse reduce_axis error in pooling schedule (#482)[BASE] Make macro namespace angostic (#480)Packing and data layout change added to conv2d_nchw (#479)add pool (#478)[TOPI] Fix reduce fusion with more levels (#477)[TOPI] dense API to remove redudant use_bias (#476)[TOPI] Fix reduction fusion with injective input (#475)[TOPI] migrate global_avg_pool, fully_connected (#472)[TEST] rfactor+ewise, cite rfactor paper (#474)[TOPI] Formalize the tag system (#473)[INTRIN] Enable pow (#471)[TOPI] negation->negative to be consistent with numpy (#470)[CODEGEN] Fix Metal codegen when storage scope is needed (#469)fix cpp deploy (#468)