Kuohais
/
GPUCodeForces
forked from ccf-ai-infra/GPUCodeForces
0
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
616 Commits 6 Branches 0 Tags 152 MiB
Commit Graph

616 Commits

Author SHA1 Message Date
Kuohais 32c35baabd Merge pull request 'finish Variance #13' (#174) from uucoco/GPUCodeForces:uucoco13 into main 2025-11-21 14:01:17 +08:00
Kuohais c3620c0263 Merge pull request 'feat: Add high-performance LayerNorm operator with vectorized Welford algorithm #1' (#175) from wut0n/GPUCodeForces:layerNorm-welford into main 2025-11-21 14:00:41 +08:00
Kuohais 26a3f5592b Merge pull request 'feat: Optimize Focal Loss with Custom CUDA Kernel #2' (#178) from wut0n/GPUCodeForces:focalloss1 into main 2025-11-21 14:00:24 +08:00
Kuohais 88252601d7 Merge pull request 'feat: Add CUDA-optimized RMSNorm implementation' (#179) from wut0n/GPUCodeForces:Rmsnorm into main 2025-11-21 14:00:13 +08:00
Kuohais bc2364209a Merge pull request 'feat: Add CUDA-optimized InstanceNorm2d implementation' (#180) from wut0n/GPUCodeForces:Instancenorm into main 2025-11-21 14:00:04 +08:00
Kuohais 851eff1f18 Merge pull request 'feat: add high-performance GroupNorm CUDA implementation #5' (#181) from wut0n/GPUCodeForces:Groupnorm into main 2025-11-21 08:58:48 +08:00
wut0n ba818537dd feat: add high-performance GroupNorm CUDA implementation 2025-11-21 00:50:30 +08:00
wut0n 773660510c feat:Instancenorm #4 2025-11-20 21:15:22 +08:00
wut0n abe2f25f6f feat:Rmsnorm #3 2025-11-20 21:12:55 +08:00
wut0n e814db7d0d feat:add focal loss optimization 2025-11-20 18:35:27 +08:00
wut0n 005be93d7f feat: Add high-performance LayerNorm operator with vectorized Welford algorithm #1 2025-11-19 21:21:51 +08:00
uucoco 25a61d5413 finish Variance #13 2025-11-18 22:01:36 +08:00
uucoco 159c077a0f finish CoVariance #12 2025-11-18 22:00:30 +08:00
uucoco 524465156a finish IntraClassCorrelation #11 2025-11-18 21:56:49 +08:00
uucoco 1e08cf9fd6 finish CharbonnierLoss #10 2025-11-18 21:54:57 +08:00
uucoco c2d6c1b0e0 finish IntraClassCorrelation #11 2025-11-18 21:50:54 +08:00
uucoco c5c4947af3 finish CharbonnierLoss #10 2025-11-18 21:49:29 +08:00
Ljy123 7f566daf34 finish rmsnorm 2025-11-18 21:02:51 +08:00
Ljy123 81359222cd finish swish 2025-11-18 21:00:38 +08:00
gsd e2487c3424 finish WelschLoss #20 2025-11-18 16:55:28 +08:00
gsd b3645c9432 finish Generator #19 2025-11-18 16:54:17 +08:00
gsd cf876ec92a finish FocalLoss #18 2025-11-18 16:52:46 +08:00
gsd 46993d4d96 finish DiceLoss #17 2025-11-18 16:51:32 +08:00
hli28146 d10b395fcf finish lppool1d #6 2025-11-18 10:35:21 +08:00
Kuohais 0d62411780 Merge pull request 'finish cholesky #7' (#126) from hli28146/GPUCodeForces:cholesky into main 2025-11-18 10:24:16 +08:00
Kuohais 4b6b3f4c3f Merge pull request 'finish ChannelShuffle #8' (#127) from uucoco/GPUCodeForces:uucoco8 into main 2025-11-18 10:24:06 +08:00
Kuohais 53c068614f Merge pull request 'finish JaccardSimilarity #9' (#128) from uucoco/GPUCodeForces:uucoco9 into main 2025-11-18 10:23:56 +08:00
Kuohais f6b18e69b9 Merge pull request 'finish CauchyLoss #16' (#137) from gsd123/GPUCodeForces:gsd16 into main 2025-11-18 10:23:32 +08:00
Kuohais 7eb66190fa Merge pull request 'finish tripletmarginloss #1' (#138) from hli28146/GPUCodeForces:h1 into main 2025-11-18 10:23:23 +08:00
Kuohais 6f62954235 Merge pull request 'finish multimarginloss #2' (#139) from hli28146/GPUCodeForces:h2 into main 2025-11-18 10:23:13 +08:00
Kuohais 614cad9fdd Merge pull request 'finish huberloss #4' (#141) from hli28146/GPUCodeForces:h4 into main 2025-11-18 10:23:04 +08:00
Kuohais 817f82808d Merge pull request 'finish softmarginloss #3' (#140) from hli28146/GPUCodeForces:h3 into main 2025-11-18 10:22:54 +08:00
Kuohais 81cae82ec7 Merge pull request 'finish multilabelmarginloss #5' (#142) from hli28146/GPUCodeForces:h5 into main 2025-11-18 10:22:38 +08:00
Kuohais 3d91fef946 Merge pull request 'finish cosineloss Operator #1' (#144) from ZZZJ/GPUCodeForces:cosineloss_opt into main 2025-11-18 10:22:21 +08:00
Kuohais db542d847f Merge pull request 'finish rope #1' (#145) from wwmm/GPUCodeForces:rope into main 2025-11-18 10:22:00 +08:00
Kuohais e897113c94 Merge pull request 'finish cosinesimilarity #2' (#147) from wwmm/GPUCodeForces:cosinesimilarity into main 2025-11-18 10:21:40 +08:00
Kuohais cefbbe0b95 Merge pull request 'finish embeddingbag #3' (#149) from wwmm/GPUCodeForces:embeddingbag into main 2025-11-18 10:21:13 +08:00
Kuohais 26d3ebff81 Merge pull request 'finish gelu_dropout operator#3' (#152) from Ljy123/GPUCodeForces:gelu into main 2025-11-18 10:20:38 +08:00
Kuohais eb2dbfb559 Merge pull request 'optimized softmax #6' (#155) from Ljy123/GPUCodeForces:softmax into main 2025-11-18 10:20:24 +08:00
Kuohais 1361864229 Merge pull request 'finish vecdot #8' (#156) from hli28146/GPUCodeForces:h8 into main 2025-11-18 10:19:43 +08:00
Kuohais 5aac06b5d9 Merge pull request 'finish cross #9' (#157) from hli28146/GPUCodeForces:h9 into main 2025-11-18 10:18:06 +08:00
Kuohais 7569b354fd Merge pull request 'finish logsumexp #10' (#158) from hli28146/GPUCodeForces:h10 into main 2025-11-18 10:17:29 +08:00
hli28146 68193644db finish logsumexp #10 2025-11-17 22:36:15 +08:00
hli28146 5650da6b59 finish cross #9 2025-11-17 22:30:08 +08:00
hli28146 3240860cca finish vecdot #8 2025-11-17 22:28:01 +08:00
Ljy123 6d4d815d0c optimzied softmax 2025-11-17 22:01:20 +08:00
Ljy123 e53b99bec3 finish gelu 2025-11-17 21:41:40 +08:00
wwmm e6591b7e23 fixes embeddingbag #3 2025-11-17 21:12:52 +08:00
wwmm b1138a4868 fixes cosinesimilarity #2 2025-11-17 21:06:28 +08:00
wwmm 2569ba9f1b fixes rope #1 2025-11-17 20:58:14 +08:00