Commit graph

  • ed391e740c
    Merge pull request #89 from lucasdelimanogueira/tmp main lucasdelimanogueira 2024-06-05 21:09:55 -03:00
  • 53983be40a v0.0.7 improve memory usage lucasdelimanogueira 2024-06-05 21:07:07 -03:00
  • 5884dff3bf fix memory usage lucasdelimanogueira 2024-06-05 21:04:32 -03:00
  • 70cbd45ae7 fix memory usage lucasdelimanogueira 2024-06-05 21:04:09 -03:00
  • f0f8641d54
    Merge pull request #88 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-06-05 20:25:50 -03:00
  • 7ac5fbcc47 fix memory usage lucasdelimanogueira 2024-06-05 20:25:16 -03:00
  • e6d58121f1 fix memory usage lucasdelimanogueira 2024-06-05 20:24:09 -03:00
  • 1bfea7544c
    Merge pull request #87 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-06-05 16:05:15 -03:00
  • e670559fe4 fix memory issues del tensors lucasdelimanogueira 2024-06-05 15:43:19 -03:00
  • 6701179c22 del tensor free memory lucasdelimanogueira 2024-06-05 14:31:48 -03:00
  • b56ed81cd7 add comment on reduce cuda lucasdelimanogueira 2024-06-04 11:21:56 -03:00
  • 195b306941
    Merge pull request #86 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-06-03 15:39:21 -03:00
  • 12eb38a940 multigpu training support v0.0.5 lucasdelimanogueira 2024-06-03 15:38:16 -03:00
  • 64f08a8963 multigpu training support v0.0.5 lucasdelimanogueira 2024-06-03 15:37:57 -03:00
  • e43928ff4b
    Merge pull request #85 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-06-03 15:16:07 -03:00
  • 7691cbc2dc fix allreduce distributed not working after 2nd iteration lucasdelimanogueira 2024-06-03 15:15:35 -03:00
  • c42d334599 fix allreduce distributed not working after 2nd iteration lucasdelimanogueira 2024-06-03 15:14:42 -03:00
  • 19e6924223
    Merge pull request #84 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-06-03 11:41:24 -03:00
  • e706b81007 fix broadcast and reduce gradients distributeddataparallel and tensor hooks lucasdelimanogueira 2024-06-03 11:40:05 -03:00
  • 8fd0ffdd65
    Merge pull request #83 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-06-01 20:57:17 -03:00
  • ef84c08841 fix backward reduce distributed lucasdelimanogueira 2024-06-01 20:55:02 -03:00
  • fcc2bedf01
    Merge pull request #82 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-06-01 10:16:10 -03:00
  • 4b683a20e9 fix distributed ncclCommInitRank lucasdelimanogueira 2024-05-31 19:40:58 -03:00
  • bfa6cceba8
    Merge pull request #81 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-31 19:04:35 -03:00
  • ef9a74b257 test cloud multigpu lucasdelimanogueira 2024-05-31 17:43:22 -03:00
  • 9735007805 test cloud multigpu lucasdelimanogueira 2024-05-31 17:42:15 -03:00
  • c8eb01ca92 test cloud multigpu lucasdelimanogueira 2024-05-31 17:37:09 -03:00
  • f09885abec test cloud multigpu lucasdelimanogueira 2024-05-31 17:36:10 -03:00
  • 5178aeabf7
    Merge pull request #80 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-31 17:16:13 -03:00
  • f3af107f33 information ami test readme p3 aws lucasdelimanogueira 2024-05-29 18:17:35 -03:00
  • 38a860eba1 fix send to other devices and compile on cloud lucasdelimanogueira 2024-05-29 17:48:37 -03:00
  • 98651d00a5
    Merge pull request #79 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-28 10:29:30 -03:00
  • 446b382333 add libs install setup lucasdelimanogueira 2024-05-28 10:28:54 -03:00
  • 13f062f89b
    Merge pull request #78 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-28 10:15:36 -03:00
  • bb88630608 fix import lib lucasdelimanogueira 2024-05-28 10:15:12 -03:00
  • c88a8fb8cc
    Merge pull request #77 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-27 21:04:28 -03:00
  • 002287d832 remove print from cuda to cpu lucasdelimanogueira 2024-05-27 20:55:50 -03:00
  • 87bf2318ca test broadcast allreduce multi gpu lucasdelimanogueira 2024-05-27 15:15:53 -03:00
  • ffe8b2f0bf
    Merge pull request #76 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-27 15:13:42 -03:00
  • 6b450a87bf fix readme lucasdelimanogueira 2024-05-27 15:13:03 -03:00
  • 100799215f fix readme lucasdelimanogueira 2024-05-27 15:11:40 -03:00
  • f5d0a1271d
    Merge pull request #75 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-27 15:10:15 -03:00
  • 240e603c0c fix distributed dataloader lucasdelimanogueira 2024-05-27 15:09:07 -03:00
  • e3bb882d3c fix distributed dataloader lucasdelimanogueira 2024-05-27 15:08:27 -03:00
  • 10a88fe8ca unit test dataloader and distributedsampler lucasdelimanogueira 2024-05-25 11:55:59 -03:00
  • e590f57555 implemented distributed sampler lucasdelimanogueira 2024-05-25 10:53:09 -03:00
  • 56fff47001 fix import data module lucasdelimanogueira 2024-05-25 09:55:02 -03:00
  • 9d7417a531 distributed sampler lucasdelimanogueira 2024-05-25 09:48:32 -03:00
  • a316dfd191 add argument src to broadcast distributed lucasdelimanogueira 2024-05-25 09:38:42 -03:00
  • 0e9c4779a8 add argument src to broadcast distributed lucasdelimanogueira 2024-05-25 09:38:14 -03:00
  • c3c8fd7199 fix name module parallel lucasdelimanogueira 2024-05-25 09:10:34 -03:00
  • 1e0c408ffb fix import distributed nn lucasdelimanogueira 2024-05-25 08:57:35 -03:00
  • 5283c1bc06 allreduce mean lucasdelimanogueira 2024-05-25 08:52:20 -03:00
  • 7a3b0f276b v1 distributed data parallel lucasdelimanogueira 2024-05-24 22:17:22 -03:00
  • 776aacabdd
    Merge pull request #74 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-24 20:53:00 -03:00
  • 01b93132bc fix setup.py installing lucasdelimanogueira 2024-05-24 20:31:31 -03:00
  • 3ed877adbf fix distributed allreduce calling python lucasdelimanogueira 2024-05-24 20:16:08 -03:00
  • 8654374a33 fix setup.py install nccl lucasdelimanogueira 2024-05-24 20:13:16 -03:00
  • 6d8ff7f9af fix distributed allreduce calling python lucasdelimanogueira 2024-05-24 20:06:34 -03:00
  • b21c7e9f34 fix setup.py lucasdelimanogueira 2024-05-24 19:57:49 -03:00
  • 5f92e6dc8c
    Merge pull request #73 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-24 19:55:49 -03:00
  • 29ee569d04 fix import distributed lucasdelimanogueira 2024-05-24 19:55:19 -03:00
  • 2720f5c88c
    Merge pull request #72 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-24 19:49:58 -03:00
  • 33d4dc14d8 fix install nccl setup.py lucasdelimanogueira 2024-05-24 19:49:25 -03:00
  • ef272b65ee fix install nccl setup.py lucasdelimanogueira 2024-05-24 19:48:37 -03:00
  • 82b6aa245f
    Merge pull request #71 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-24 18:37:11 -03:00
  • b637c3e0f1 fix readme lucasdelimanogueira 2024-05-24 18:36:17 -03:00
  • 049a515e45
    Merge pull request #70 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-24 18:35:30 -03:00
  • 94b3582a46 test broadcast and all reduce lucasdelimanogueira 2024-05-24 18:31:59 -03:00
  • 55db13836b broadcast allreduce nccl lucasdelimanogueira 2024-05-24 18:19:39 -03:00
  • e55c1195ec implemented mpi for distributed run lucasdelimanogueira 2024-05-24 15:12:53 -03:00
  • b152ae2a1d first commit distributed module lucasdelimanogueira 2024-05-23 19:15:30 -03:00
  • 15a817e295
    Merge pull request #69 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-23 12:29:04 -03:00
  • 55fc1ac020 version 0.0.4 cuda running neural network lucasdelimanogueira 2024-05-23 12:28:46 -03:00
  • 8c5abb226c version 0.0.4 cuda running neural network lucasdelimanogueira 2024-05-23 12:28:17 -03:00
  • 77cabe41a2 fix cross entropy loss cuda lucasdelimanogueira 2024-05-23 12:19:32 -03:00
  • 4676a70d14 fix max min sum access elements autograd backward running on device lucasdelimanogueira 2024-05-23 12:16:07 -03:00
  • 9b49a07d2a fix max min sum access elements autograd backward running on device lucasdelimanogueira 2024-05-23 12:11:59 -03:00
  • 27feb8a366 fix autograd was not running on device lucasdelimanogueira 2024-05-23 12:07:17 -03:00
  • 472f3189c2 fix cross entropy loss send to device lucasdelimanogueira 2024-05-23 10:41:52 -03:00
  • e8e0060955 fix cross entropy loss send to device lucasdelimanogueira 2024-05-23 10:35:55 -03:00
  • 8b19ac38c9 fix send module to device lucasdelimanogueira 2024-05-23 10:20:08 -03:00
  • f13d22eb91 fix send to device unittest lucasdelimanogueira 2024-05-23 10:04:44 -03:00
  • b1c28c4cea fix send to device lucasdelimanogueira 2024-05-23 10:00:48 -03:00
  • d7a932f8ed fix send to device lucasdelimanogueira 2024-05-23 09:48:16 -03:00
  • 065e913b2d fix send to device lucasdelimanogueira 2024-05-23 09:46:35 -03:00
  • a072b57b50 fix module send to cuda lucasdelimanogueira 2024-05-23 03:19:49 -03:00
  • 374792af81
    Merge pull request #68 from lucasdelimanogueira/tmp lucasdelimanogueira 2024-05-23 03:15:52 -03:00
  • 5946e6e1bd fix unittest test nn cuda lucasdelimanogueira 2024-05-23 03:01:27 -03:00
  • d70e398773 fix autograd cuda unittests lucasdelimanogueira 2024-05-23 02:57:57 -03:00
  • fb4f90b0cf small fix 3D transpose lucasdelimanogueira 2024-05-23 02:35:00 -03:00
  • a0bfd6a0eb small fix 3D transpose lucasdelimanogueira 2024-05-23 02:34:38 -03:00
  • 38e629fc38 make contiguous cuda lucasdelimanogueira 2024-05-23 02:27:46 -03:00
  • d2fc1e3617 make contiguous cuda lucasdelimanogueira 2024-05-23 02:27:30 -03:00
  • 702ebaf87c 1D, 2D and 3D transpose cuda lucasdelimanogueira 2024-05-23 02:09:32 -03:00
  • fb6429aad6 min cuda and min cuda axis lucasdelimanogueira 2024-05-23 01:39:37 -03:00
  • 9708054ecd min cuda and min cuda axis lucasdelimanogueira 2024-05-23 01:36:04 -03:00
  • 223229fcd3 min cuda and min cuda axis lucasdelimanogueira 2024-05-23 01:35:45 -03:00
  • e2a0262c4e small fix max cuda axis was setting init to 0 instead of -inf lucasdelimanogueira 2024-05-23 01:29:23 -03:00
  • 5882f1d801 max cuda and max cuda axis lucasdelimanogueira 2024-05-23 01:26:28 -03:00