Skip to content
Snippets Groups Projects
Select Git revision
  • summer default
1 result
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.027Nov262524232221201918171615131211987653231Oct30292827262322212019181716151413121098765432130Sep29282725242322171615141211109876543231Aug30272524232221201918171614131211108765432131Jul3029282726252423222120191817161514131211109876542130Jun2928272625242322211918171615141312111098fix distribute_clone sbp (#3803)New checkpoint (#3540)Reshape backward issue with distribute split (#3915)bump 0.3b1Remove NormalModelUpdateOpConf (#3917)Add triplet loss (#3864)Dev unsorted segment sum (#3731)fix include files not copied (#3907)Dev split like add backward (#3901)optimize cfg generator to save time (#3906)Add polyval op (#3541)DynamicLossScale (#3886)add mseloss (#3893)distribute concat out dynamic false (#3899)add HasGradTensor4OpOutput (#3904)fix gather out set is_dynamic (#3900)fix unsorted_segment_sum (#3898)add check in deconv (#3835)add bceloss (#3804)Add ssp variable proxy (#3859)Op collection (#3833)multi_count_not_finite op (#3879)model_update_op_add_skip_if (#3883)Simple kernel memory bandwidth profiler (#3855)rm scope_proto symbol_id (#3865)fix stirng format in py35 (#3878)optimize rmsprop and lars optimizers (#3809)fix pybind11 include (#3876)Dev enhance sort ops (#3828)Dev replace str to cfg obj in python callback (#3832)fix CUDA_NVCC_GENCODES (#3869)Cfg template operator reform (#3861)Dev use union instead of struct for oneof (#3870)Py kernel2 (#3686)add oneflow_range (#3725)SplitLikeOp support prefix (#3866)CI merge all jobs into one (#3868)add l1 loss op in python (#3793)Optimize softmax cuda kernel block size (#3853)Check label every push (#3863)