Commit Graph

  • d8aaed35cb
    Merge 2f11a94dd9 into 65bf82ffce Rifky Bujana Bisri 2023-08-25 22:31:10 +0000
  • 2f11a94dd9
    Fix error cublasLt not found on Cuda version 11 and up with Ubuntu 18.0+ Rifky Bujana Bisri 2023-08-25 15:22:00 -0700
  • a3d85da11a
    Merge 9f22802d87 into 65bf82ffce dependabot[bot] 2023-06-07 15:03:29 +0000
  • 9f22802d87
    Bump src/3rd_party/fbgemm from 6f45243 to 0e33146 dependabot/submodules/src/3rd_party/fbgemm-0e33146 dependabot[bot] 2023-06-07 15:03:27 +0000
  • c6e768e12d
    Merge a45cc0361c into 65bf82ffce dependabot[bot] 2023-04-10 15:07:48 +0000
  • a45cc0361c
    Bump examples from 58f48a0 to 6c40475 dependabot/submodules/examples-6c40475 dependabot[bot] 2023-04-10 15:07:47 +0000
  • 01f08dbcd9
    Merge f395ebd480 into 65bf82ffce dependabot[bot] 2023-03-30 15:06:47 +0000
  • f395ebd480
    Bump regression-tests from 2a8bed3 to 89ce02e dependabot/submodules/regression-tests-89ce02e dependabot[bot] 2023-03-30 15:06:44 +0000
  • a67e70e52b
    Bump regression-tests from 2a8bed3 to 0ddf105 dependabot[bot] 2023-03-28 15:12:26 +0000
  • 65bf82ffce
    version 1.12.0 (#980) master 1.12.0 Marcin Junczys-Dowmunt 2023-02-21 09:56:29 -0800
  • 4ffd292881 Merge branch 'master' into pmaster Marcin Junczys-Dowmunt 2023-02-20 12:15:33 -0800
  • 031dbb3266 Merged PR 27804: Fallback to old LSH code for MSVC due to bad loop unrolling Marcin Junczys-Dowmunt 2023-02-13 15:44:19 +0000
  • 9ad5203ca2 Merged PR 26476: Sanitize guided-alignment with case-augmentation (still somewhat wonky) Marcin Junczys-Dowmunt 2023-02-11 16:35:29 +0000
  • 4f145c450f Merged PR 26311: [FSM] make model loading lock non-static Varun Mathur 2023-02-10 16:34:37 +0000
  • ee50d4aaea Merged PR 27051: Add an option for completely resetting validation metrics Roman Grundkiewicz 2022-12-20 17:56:10 +0000
  • b7205fc0b0 Merged PR 25220: Add extra model information to model_info.py script Alex Muzio 2022-11-30 12:23:38 +0000
  • 3c2a432995
    Bump examples from 25e8438 to 58f48a0 (#975) dependabot[bot] 2022-11-29 18:22:22 +0000
  • d5569cee60
    Bump regression-tests from 494d6de to 488d454 (#974) dependabot[bot] 2022-11-29 18:22:00 +0000
  • b6581c4c44 Merged PR 26667: Update examples submodule to fix vulnerability issues Roman Grundkiewicz 2022-11-23 19:16:44 +0000
  • c79dc80a2f Merged PR 26617: Update regression-tests & fix CI pipelines Roman Grundkiewicz 2022-11-20 13:31:10 +0000
  • 36349645b8
    Bump src/3rd_party/sentencepiece from 31ac8e8 to 8dc9172 (#970) dependabot[bot] 2022-11-19 08:36:06 +0000
  • 4187aab9ba
    Bump regression-tests from 92e116e to 494d6de (#973) dependabot[bot] 2022-11-19 08:35:45 +0000
  • 07a2ac8126
    best-deep alias broken (#968) Nikolay Bogoychev 2022-11-02 11:16:14 +0000
  • cda2f2112f
    Temporarily download MKL tarball from a mirror server (#972) Roman Grundkiewicz 2022-11-02 11:09:43 +0000
  • be1ee3fa94 Merged PR 26318: Fix incorrect envvar name in Azure Pipeline Roman Grundkiewicz 2022-11-01 10:07:40 +0000
  • a6de1b781c Merged PR 26271: Update CI pipeline triggers Roman Grundkiewicz 2022-11-01 06:26:56 +0000
  • 4d3702c4ec Merged PR 25950: Add missing defaults for concatenated factors Marcin Junczys-Dowmunt 2022-10-06 05:53:16 +0000
  • da6e30bfe3 merge with internal master Marcin Junczys-Dowmunt 2022-10-03 17:51:03 -0700
  • 1e92cff93d Merged PR 25919: Sync with public master - no review required Marcin Junczys-Dowmunt 2022-10-04 00:42:52 +0000
  • 2c55cdb3c0 Merged PR 25889: Fixes bad memory access problem in hashing Marcin Junczys-Dowmunt 2022-09-29 19:01:49 +0000
  • 2cd3055d76 Merged PR 25836: Check via hashing if re-syncing in local mode is required Marcin Junczys-Dowmunt 2022-09-27 18:40:53 +0000
  • 1f2929d528 Merged PR 25733: Fused inplace ReLU and Dropout in transformer FFN layer Marcin Junczys-Dowmunt 2022-09-26 20:17:33 +0000
  • cfc33f5498 only use tcmalloc_minimal Marcin Junczys-Dowmunt 2022-09-22 15:11:33 -0700
  • 7d2045a907 Merged PR 25686: Loading checkpoints from main node only via MPI Marcin Junczys-Dowmunt 2022-09-21 20:39:54 +0000
  • 76964791ad Merged PR 23767: More principled sampling and force-decoding Marcin Junczys-Dowmunt 2022-09-16 22:53:08 +0000
  • e13053a6f2 Merged PR 25698: Install Python 3.8 on GPU pool Roman Grundkiewicz 2022-09-16 09:30:10 +0000
  • 6f7766f837 Merged PR 25465: Choose top checkpoints from train.log for averaging Roman Grundkiewicz 2022-09-15 06:19:18 +0000
  • a47912d9f1 Merged PR 25518: Upgrade Azure Pipelines to macos-12 Roman Grundkiewicz 2022-09-15 06:18:42 +0000
  • 6b41df2a44 Version 1.11.8 Roman Grundkiewicz 2022-09-05 16:53:55 +0100
  • a5223e28ae
    Bump examples from 29f4f7c to 25e8438 (#963) dependabot[bot] 2022-09-05 16:52:36 +0100
  • b6d066794e
    Bump regression-tests from 4fa9ff5 to 92e116e (#964) dependabot[bot] 2022-09-05 16:52:16 +0100
  • 347ab4d54a
    Upgrade dependencies in the documentation framework (#965) Roman Grundkiewicz 2022-09-05 16:51:48 +0100
  • 0afe2478aa
    Upgrade workflows to ubuntu-20.04 and macos-12 (#962) Roman Grundkiewicz 2022-09-02 16:30:45 +0100
  • 7d654603bf
    Fix guaranteed YAML::InvalidNode when compiled with COMPILE_CPU=Off (#944) Jelmer 2022-09-02 12:25:42 +0100
  • bf5eafab0c
    Bump src/3rd_party/intgemm from a05a2e5 to 0eda93a (#933) dependabot[bot] 2022-09-02 12:22:54 +0100
  • 650cf19e5a
    Update Catch2 from 2.10.1 to 2.13.9 (#941) Graeme Nail 2022-09-02 10:36:43 +0100
  • 3bd281c6c9
    Fix clang 13.0.1 (#939) Nikolay Bogoychev 2022-09-02 10:36:15 +0100
  • 6250cd88bb
    Fixed some warnings on clang 15 that are promoted into errors (#936) KOLANICH 2022-09-02 08:04:23 +0000
  • f9a1ed10ce
    Add a workflow compiling Marian using clang-14 (#940) Roman Grundkiewicz 2022-09-02 07:19:33 +0100
  • 5d466bc367 Merged PR 25507: Upgrade Azure Pipelines to ubuntu-20.04 Roman Grundkiewicz 2022-09-02 05:55:20 +0000
  • a90950ea25 Merged PR 25154: Add model shapes flag to model_info.py script Alex Muzio 2022-08-10 22:23:47 +0000
  • e27da62393
    Directory listing in Ubuntu and macOS workflows (#938) Graeme Nail 2022-06-06 13:32:58 +0100
  • c5081df93f Merged PR 24111: Remove external reference to Docker images Roman Grundkiewicz 2022-05-31 15:31:39 +0000
  • 5df240f534
    Update status badges (#935) Roman Grundkiewicz 2022-05-31 12:38:47 +0100
  • 042ed8f2e2 Merged PR 24072: Revert changes to transformer caching Marcin Junczys-Dowmunt 2022-05-30 07:27:15 +0000
  • f3e1efe731 merge with internal master Marcin Junczys-Dowmunt 2022-05-26 06:28:06 -0700
  • 95720ae19f
    Update NVIDIA CUDA signing key for CI; fix for building docs (#932) Graeme Nail 2022-05-18 11:11:28 +0100
  • 704a323142 Merged PR 22799: Running regression tests on Azure Pipelines Roman Grundkiewicz 2022-05-13 07:30:36 +0000
  • e0e3287a3b Merged PR 23840: Update CUDA installation script for Ubuntu Roman Grundkiewicz 2022-05-12 16:23:58 +0000
  • e4f3d0f740 add fallback option for sampling, for back-compat Marcin Junczys-Dowmunt 2022-05-09 13:28:28 -0700
  • 1a74358277 Merged PR 23429: Small fixes around fp16 training and batch fitting Marcin Junczys-Dowmunt 2022-04-11 20:19:58 +0000
  • 1e4e1014ed Merged PR 23415: Set Windows image back to windows-2019 Roman Grundkiewicz 2022-04-08 17:15:56 +0000
  • d5c7372a67 Merged PR 23407: Fix incorrect/missing gradient accumulation for affine biases Marcin Junczys-Dowmunt 2022-04-08 16:00:04 +0000
  • 23c36ec1a3
    Fixed fp16 training/inference with factors-combine concat (#926) Artur Nowakowski 2022-03-22 11:07:41 +0100
  • 78bef7aeba
    Bump src/3rd_party/sentencepiece from c307b87 to 5312a30 (#927) dependabot[bot] 2022-03-22 10:06:11 +0000
  • 75a7a1dfd2
    Bump regression-tests from 88e6382 to 4fa9ff5 (#929) dependabot[bot] 2022-03-22 08:40:11 +0000
  • c809843f14
    Bump examples from 6d5921c to 29f4f7c (#928) dependabot[bot] 2022-03-22 08:38:30 +0000
  • 16bfa0c913 Merged PR 23094: Adapt --cost-scaling to more stable setting Marcin Junczys-Dowmunt 2022-03-16 14:44:17 +0000
  • 310d2f42f6 Merged PR 22939: Fix case augmentation with multi-threaded reading Marcin Junczys-Dowmunt 2022-03-07 16:57:32 +0000
  • adaaf087e4 better error message Marcin Junczys-Dowmunt 2022-02-16 13:20:48 -0800
  • 601c9ac980
    Detect fortran_order in npz (#911) Graeme Nail 2022-02-15 13:22:49 +0000
  • 58c4576e5d
    Bump regression-tests from da95717 to 88e6382 (#923) dependabot[bot] 2022-02-15 11:21:14 +0000
  • 8a9580b329
    update the intgemm version to upstream (#924) Nikolay Bogoychev 2022-02-15 11:18:29 +0000
  • b8bf086b10 move regression-tests pointer Marcin Junczys-Dowmunt 2022-02-11 06:04:38 -0800
  • b0275e7754 merge with internal master Marcin Junczys-Dowmunt 2022-02-11 06:03:16 -0800
  • 4b51dcbd06 Merged PR 22524: Optimize guided alignment training speed via sparse alignments - part 1 Marcin Junczys-Dowmunt 2022-02-11 13:50:47 +0000
  • 3b21ff39c5 update VERSION and CHANGELOG Marcin Junczys-Dowmunt 2022-02-10 08:35:49 -0800
  • b3feecc82b Merged PR 22483: Make C++17 the official standard for Marian Marcin Junczys-Dowmunt 2022-02-10 16:34:23 +0000
  • e6dbacb310 Merged PR 22490: Faster LSH top-k for CPU Marcin Junczys-Dowmunt 2022-02-10 16:30:21 +0000
  • 8fd553e582
    Bump examples from 6d5921c to 0ca966e (#919) dependabot[bot] 2022-02-10 14:03:37 +0000
  • 17e55f5a7d
    Update VERSION Roman Grundkiewicz 2022-02-10 11:20:47 +0000
  • 4d44627f26
    PyYaml safe_load instead of load (#913) Graeme Nail 2022-02-10 11:20:27 +0000
  • a492bc57d2
    Bump regression-tests from 0716f4e to f7971b7 (#918) dependabot[bot] 2022-02-10 10:28:04 +0000
  • 73f1899307
    Add dependabot for git submodules (#916) Roman Grundkiewicz 2022-02-10 10:25:08 +0000
  • b97645846a
    Update release workflow (#915) Roman Grundkiewicz 2022-02-09 18:56:56 +0000
  • bcf29b8cd2
    Update acknowledgements (#914) Graeme Nail 2022-02-09 17:05:48 +0000
  • f00d062189 update VERSION and CHANGELOG - Release 1.11.0 1.11.0 Marcin Junczys-Dowmunt 2022-02-08 08:39:24 -0800
  • 8e659bb5c0
    Document Structure (#910) Graeme Nail 2022-02-08 10:58:09 +0000
  • 05ba9e4c31
    add -DDETERMINISTIC=ON/OFF flag (#912) Marcin Junczys-Dowmunt 2022-02-08 02:57:20 -0800
  • a365bb5ce9 fix server behaviour Marcin Junczys-Dowmunt 2022-02-07 08:09:54 -0800
  • aafe8fb5ca update regression tests pointer Marcin Junczys-Dowmunt 2022-02-07 02:36:20 -0800
  • 3cf9e83bac resolve conflicts Marcin Junczys-Dowmunt 2022-02-06 12:33:58 -0800
  • 8da539e835 merged with master Marcin Junczys-Dowmunt 2022-02-06 12:00:48 -0800
  • 266b931daa
    Update list of contributors (#906) Roman Grundkiewicz 2022-01-30 20:11:38 +0000
  • 07c39c7d76
    Cherry picked cleaning/refeactoring patches (#905) Roman Grundkiewicz 2022-01-28 14:16:41 +0000
  • 71b5454b9e
    Layer documentation (#892) Qianqian Zhu 2022-01-26 15:17:38 +0000
  • 3b458b044e
    Update VERSION Roman Grundkiewicz 2022-01-24 15:28:37 +0000
  • 894a07ad5b
    Improve checks on transformer cache (#881) Graeme Nail 2022-01-24 15:28:13 +0000
  • b64e258bda
    Update VERSION Roman Grundkiewicz 2022-01-18 12:59:37 +0000
  • b29cc07a95
    Scorer model loading (#860) Graeme Nail 2022-01-18 12:58:52 +0000