v1.2.0

What's Changed

• Add typeguard dependency for llama2 v5.0 Nvidia implementation by @anandhu-eng in #627
• Base code changes for v5.1 by @anandhu-eng in #626
• Add support for older mlcflow versions by @arjunsuresh in #628
• Support libcxx for install,llvm,src by @arjunsuresh in #630
• Fix libcxx runtime for install-llvm by @arjunsuresh in #631
• Include libunwind for install-llvm-src by @arjunsuresh in #632
• Support service account download in R2 by @anandhu-eng in #634
• Add input mapping for downloading with service account credentials by @anandhu-eng in #637
• Fixes for llama2 by @anandhu-eng in #639
• Support more libs by @arjunsuresh in #641
• Update document-scripts.yml by @arjunsuresh in #642
• Fix gfortran deps for flang cross-compilation by @arjunsuresh in #643
• Map the system name env variable by @anandhu-eng in #645
• Prevent applying patch for nvmitten in v5.1 by @anandhu-eng in #644
• Add description to args for downloading model and dataset in host by @anandhu-eng in #646
• Add support for cache expiration in get-cuda-devices by @anandhu-eng in #647
• Fix issue with temp file paths by @anandhu-eng in #648
• Support update_meta_if_env at script level by @amd-arsuresh in #649
• Added support for google-generativeai, groq, pdfplumber, and python-dotenv in meta.yaml for LLM-Evaluation by @sujik18 in #651
• Fix bug in docker-run logging by @arjunsuresh in #654
• Merge Dev by @arjunsuresh in #650
• Skip ownership restore - Waymo dataset extraction by @anandhu-eng in #655
• Fix #659, hpcx_paths usage in nvidia implementation by @arjunsuresh in #660
• Support r2-downloader for llama3.1-405b asset downloads by @anandhu-eng in #658
• Support RGAT model download via r2-downloader by @anandhu-eng in #657
• Improvements for install-llvm script, cache expiration by @arjunsuresh in #663
• Support docker privileged as input argument by @anandhu-eng in #661
• Support r2-downloader for nuscenes dataset download by @anandhu-eng in #667
• Changes to support the new compliance directory structure by @anandhu-eng in #666
• Support ssd model download through r2-downloader by @anandhu-eng in #669
• Support r2 downloader for deeplabv3+ by @anandhu-eng in #670
• Use r2-downloader for cognata dataset by @anandhu-eng in #668
• Support remote-run script action (WIP) by @arjunsuresh in #673
• Fixes for remote-run script by @arjunsuresh in #674
• Support deepseek-r1 model download through r2-downloader by @anandhu-eng in #675
• Support r2-downloader for dlrm model download script by @anandhu-eng in #676
• Support r2-downloader for preprocessed criteo by @anandhu-eng in #677
• Support Mixtral model and openorca download through r2 downloader by @anandhu-eng in #678
• Support r2 downloader for mixtral dataset by @anandhu-eng in #679
• Add support for adding r2-downoader for GPT-j by @anandhu-eng in #680
• Fix for openorca nvidia preprocessing by @anandhu-eng in #686
• support r2-downloader for waymo calibration dataset download by @anandhu-eng in #681
• Add tests for get-dataset-waymo by @anandhu-eng in #682
• add tests by @anandhu-eng in #683
• Support r2 downloader for sdxl model download by @anandhu-eng in #684
• Update tags in meta.yaml for GPT-J raw model by @anandhu-eng in #685
• Changes in meta for migrating to R2 by @anandhu-eng in #687
• Fixes for llvm installation from src by @arjunsuresh in #690
• Refactor update_state_from_variations, support dynamic variation in combination of variations by @amd-arsuresh in #691
• Merge docker env changes by @amd-arsuresh in #695
• Fix --action not working for docker runs by @arjunsuresh in #696
• Merge from AMD by @arjunsuresh in #697
• Refactor postprocess function - including condition for dry run by @anandhu-eng in #698
• commit for generating readme for get-sut-description by @anandhu-eng in #699
• Fix PATH export for get-aocc by @arjunsuresh in #700
• Added Intel PIN tool by @amd-arsuresh in #701
• Fixes for Intel SDE and PIN tools by @arjunsuresh in #702
• Merge from AMD by @amd-arsuresh in #704
• Merge from AMD by @amd-arsuresh in #705
• Merge from AMD by @amd-arsuresh in #706
• Merge from AMD by @arjunsuresh in #707
• Changes to support new submission directory structure by @anandhu-eng in #689

Full Changelog: v1.1.0...v1.2.0

v1.1.0

What's Changed

• Merge Dev by @arjunsuresh in #296
• Fix postfix deps by @arjunsuresh in #321
• Merge Dev by @arjunsuresh in #320
• Fixes for aocc by @arjunsuresh in #324
• Modularize dmidecode meminfo parsing by @arjunsuresh in #325
• Support aocc download and install by @arjunsuresh in #326
• Sync Dev by @arjunsuresh in #327
• Add tests for mlc-scripts installation by @anandhu-eng in #323
• Fix AOCC install by @arjunsuresh in #329
• Support Intel SDE Tool, improved individual script tests by @arjunsuresh in #333
• Update meta.yaml by @arjunsuresh in #334
• Added tests to the generic script template by @arjunsuresh in #335
• Support flang in llvm-install by @arjunsuresh in #336
• Add option to skip the certificate check for rclone downloads by @anandhu-eng in #331
• Fixed meta for get-aocc by @arjunsuresh in #337
• Sync Dev by @arjunsuresh in #330
• Verify SSL moved to Script Automation by @anandhu-eng in #338
• Add dry run + generalise rclone download by @anandhu-eng in #339
• Change logging levels based on verbose and silent by @anandhu-eng in #305
• link submit-mlperf-results with generate-mlperf-inference-submission by @anandhu-eng in #340
• Handle submission base directory properly by @anandhu-eng in #342
• Support relative path in docker run by @anandhu-eng in #343
• Added kill process script by @arjunsuresh in #347
• Support base variation inside dynamic variations by @arjunsuresh in #348
• Fix for llvm version handling by @arjunsuresh in #350
• Fix for llvm version handling by @arjunsuresh in #351
• Fixes for Nvidia gptj by @arjunsuresh in #352
• Handle situation when cache is not present by @anandhu-eng in #356
• Merge from GO by @arjunsuresh in #358
• Support google-dns for nvidia gpt docker by @arjunsuresh in #361
• Fixes nvidia gptj model generation by @arjunsuresh in #362
• Fix for dmidecode + support mounting filepaths in docker by @anandhu-eng in #344
• Replace boolean usage by @anandhu-eng in #357
• Sync Dev by @arjunsuresh in #341
• Replace KSM with 1Password in test-mlperf-inference-tvm-resnet50.yml by @nathanw-mlc in #359
• Sync main by @arjunsuresh in #363
• Fix invalid format output error in fetch-secret job. by @nathanw-mlc in #366
• Support OpenAI API by @arjunsuresh in #368
• Added run files for openai call by @arjunsuresh in #369
• Sync main with dev by @anandhu-eng in #370
• Skip authentication when service account credentials are provided by @anandhu-eng in #371
• Fix import for OpenOrca by @anandhu-eng in #375
• Fix dgl version for mlperf inference rgat by @arjunsuresh in #377
• Path string fix by @anandhu-eng in #376
• Fix command generation by @anandhu-eng in #379
• fix command generation by @anandhu-eng in #380
• Fix command generation - paths with space by @anandhu-eng in #381
• Fix for handling space by @anandhu-eng in #383
• fixes for path issues by @anandhu-eng in #384
• Path str fix by @anandhu-eng in #385
• fix for path issue by @anandhu-eng in #386
• Fix for space in path by @anandhu-eng in #387
• Fixes the output path when there is space - compiler linkage by @anandhu-eng in #388
• Fix space in path issue for dump freeze by @anandhu-eng in #389
• Run benchmark with forked inference repo by @anandhu-eng in #390
• Corrected git repo link by @anandhu-eng in #391
• Fix for space in path - get generic python lib by @anandhu-eng in #392
• Fixes for R50 cpp GH action by @arjunsuresh in #394
• Update test-amd-mlperf-inference-implementations.yml by @arjunsuresh in #395
• Replace print with MLC Logger by @anandhu-eng in #396
• Use num_threads=1 for retinanet by @arjunsuresh in #397
• Added experiment to script automation by @anandhu-eng in #398
• add --multi-thread-streams=0 for rclone version >= 1.60.0 by @anandhu-eng in #402
• Fixes for llvm-install-src by @arjunsuresh in #404
• Sync Dev by @arjunsuresh in #378
• Improvements for install-gcc-src by @arjunsuresh in #405
• Contest 2025 gemini call by @H9660 in #403
• Fixes for oneapi by @arjunsuresh in #408
• Merge from GO by @arjunsuresh in #409
• Support mlc experiment script by @arjunsuresh in #411
• Support mlc experiment entries by @arjunsuresh in #412
• Support state info for experiment run by @arjunsuresh in #413
• Support exp_tags for experiments by @arjunsuresh in #414
• Fix nvidia-dali version for python3.8, fixes #410 by @arjunsuresh in #415
• Force str for version in script module by @arjunsuresh in #416
• Code changes for integrating nvidia v5.0 by @anandhu-eng in #417
• PyCuda version fix by @anandhu-eng in #418
• Fix for pycuda in nvidia-impl by @arjunsuresh in #420
• Deprecated version in app-mlperf-inference-nvidia-scripts by @anandhu-eng in #421
• Fixes for pycuda and versions by @anandhu-eng in #422
• Removed pycuda version fix by @arjunsuresh in #423
• Fix onnx version by @anandhu-eng in #425
• Fixes to docker mounts/user by @arjunsuresh in #426
• Update customize.py by @arjunsuresh in #427
• Support docker_build_env, fixes #424 by @arjunsuresh in #428
• Fix env export for get-mlperf-inference-src by @arjunsuresh in #430
• Fix ucx LD_LIBRARY_PATH for app-mlperf-inference-nvidia by @arjunsuresh in #432
• Added set-cpu-freq script by @arjunsuresh in #435
• Added get-lib-jemalloc by @arjunsuresh in #437
• Support mlc test for scripts needing PAT in github actions by @arjunsuresh ...

v1.0.1

What's Changed

• Fix a bug in reuse_existing_container by @arjunsuresh in #227
• Prevent errors on get-platform-details by @arjunsuresh in #228
• Make noinfer-scenario results the default for mlperf-inference by @arjunsuresh in #230
• Use inference dev branch for submission preprocess by @arjunsuresh in #231
• Fix mlcr usage in docs and actions by @arjunsuresh in #232
• Fix run-mobilenet by @arjunsuresh in #233
• Fixes to run-all scripts by @arjunsuresh in #234
• Fix for issue #236 by @anandhu-eng in #237
• Refactored pointpainting model download script by @anandhu-eng in #238
• Add support for downloading waymo from mlcommons checkpoint by @anandhu-eng in #235
• Added get-aocc script by @arjunsuresh in #240
• Minor fixes to improve submission generation experience by @arjunsuresh in #242
• Make full the default variation for retinanet dataset by @anandhu-eng in #241
• Fixes for mlperf inference submissions by @arjunsuresh in #243
• Update meta.yaml | Fix protobuf version_min for R50 TF by @arjunsuresh in #244
• Exit on error - git commit by @anandhu-eng in #245
• Fixes for mlperf submission by @arjunsuresh in #249
• Cleaned the boolean usage in MLCFlow by @Sid9993 in #246
• Map rocm and gpu to cuda by @anandhu-eng in #251
• Fixes get,igbh,dataset on host by @arjunsuresh in #252
• Update customize.py | Fix boolean value for --compliance by @arjunsuresh in #254
• Fix for no-cache in run-mobilenets by @arjunsuresh in #256
• Added alternative download link for imagenet-aux by @arjunsuresh in #257
• Code cleanup by @arjunsuresh in #259
• Make low disk usage the default in mobilenet run by @arjunsuresh in #264
• Add script to download waymo calibration dataset by @anandhu-eng in #265
• Fixes for mobilenet run by @arjunsuresh in #266
• Support mlperf inference submission tar file generation by @arjunsuresh in #267
• Convert relative to abs file path by @anandhu-eng in #270
• Cleanup for run-mobilenet script by @arjunsuresh in #272
• Added command to untar waymo dataset files by @arjunsuresh in #274
• Support min_duration by @arjunsuresh in #277
• Cleanup mobilenet runs by @arjunsuresh in #279
• Update classification.cpp by @arjunsuresh in #280
• Fix duplication of automation object by @anandhu-eng in #282
• Updated mixtral dataset download based on latest inference readme by @anandhu-eng in #284
• Updated download url - llama3 by @anandhu-eng in #285
• Fix argument issue in coco2014 calibration dataset download by @anandhu-eng in #286
• Update script to detect Podman in the system by @anandhu-eng in #287
• Add get-oneapi compiler(WIP) by @anandhu-eng in #294
• Support OS_FLAVOR_LIKE and OS_TYPE in run scripts by @arjunsuresh in #297
• Fixes for wkhtmltopdf by @arjunsuresh in #299
• Fix wkhtmltopdf on macos by @arjunsuresh in #300
• Added macos install wkhtmltopdf by @arjunsuresh in #301
• Update test-mlc-script-features.yml by @arjunsuresh in #302
• Test mlc tests by @arjunsuresh in #303
• Added libgl deps for imagenet preprocessing by @arjunsuresh in #304
• Added Resnet50 closed division github action by @sujik18 in #289
• Fix for ResNet50 Closed Division GitHub Action by @sujik18 in #307
• Fix lists in install-llvm-src by @arjunsuresh in #309
• Fix wkhtmltopdf installation on windows by @arjunsuresh in #312
• Improve get-rclone-config by @arjunsuresh in #313
• Support send-email by @arjunsuresh in #317
• Support email by @arjunsuresh in #319

New Contributors

• @Sid9993 made their first contribution in #246

Full Changelog: mlperf-automations-v1.0.0...v1.0.1

mlperf-automations v1.0.0

Current release does not guarantee any reproducibility -- we are aiming it from v2.0.0

What's Changed

• Fixes for igbh dataset download by @arjunsuresh in #1
• run check-broken-links on pull request by @anandhu-eng in #2
• Code cleanup and github action added for MLPerf inference r-gat by @arjunsuresh in #49
• Capture container tool by @anandhu-eng in #50
• [Automated Commit] Format Codebase by @arjunsuresh in #51
• Fixes for rgat submission generation by @arjunsuresh in #52
• Fixes for rgat submission generation by @arjunsuresh in #53
• Updates to MLPerf inference github actions by @arjunsuresh in #54
• Support nvmitten for aarch64 by @arjunsuresh in #55
• Copy bert model for nvidia-mlperf-inference implementation instead of softlink by @arjunsuresh in #56
• Update version by @arjunsuresh in #57
• Update github actions - use master branch of inference repository by @arjunsuresh in #58
• Migrate MLPerf inference unofficial results repo to MLCommons by @arjunsuresh in #59
• Create reset-fork.yml by @arjunsuresh in #60
• Fix scc24 github action by @arjunsuresh in #61
• Fix dangling softlink issue with nvidia-mlperf-inference-bert by @arjunsuresh in #64
• Support pull_inference_changes in run-mlperf-inference-app by @arjunsuresh in #65
• Added pull_inference_changes support to run-mlperf-inference-app by @arjunsuresh in #66
• Fix github action failures by @arjunsuresh in #68
• Support --outdirname for ml models, partially fixed #63 by @sahilavaran in #71
• Update test-cm-based-submission-generation.yml by @arjunsuresh in #73
• Fix exit code for docker run failures by @arjunsuresh in #74
• Support --outdirname for datasets fixes #63 by @sahilavaran in #75
• Support version in preprocess-submission, cleanups for coco2014 script by @arjunsuresh in #76
• Fixes for nvidia-mlperf-inference by @arjunsuresh in #77
• Fix coco2014 sample ids path by @arjunsuresh in #78
• Fixes for podman support by @arjunsuresh in #79
• Not use SHELL command in CM docker by @arjunsuresh in #82
• Support adding dependent CM script commands in CM dockerfile by @arjunsuresh in #83
• Fixes for igbh dataset detection by @arjunsuresh in #85
• 2024 December Updates by @arjunsuresh in #69
• Copied mlperf automotive CM scripts by @arjunsuresh in #86
• Generated docker image name - always lower case by @anandhu-eng in #87
• Fixes for podman by @arjunsuresh in #88
• Dont use ulimit in docker extra args by @arjunsuresh in #89
• Rename ENV CM_MLPERF_PERFORMANCE_SAMPLE_COUNT by @arjunsuresh in #90
• Fix env corruption in docker run command by @arjunsuresh in #92
• Fixes for R-GAT submission generation by @arjunsuresh in #93
• Fixes for podman run, github actions by @arjunsuresh in #95
• Fix SUT name update in mlperf-inference-submission-generation by @arjunsuresh in #96
• Update format.yml by @arjunsuresh in #97
• Added submit-mlperf-results CM script for automatic mlperf result submissions by @arjunsuresh in #98
• Merge with dev by @arjunsuresh in #99
• Merge pull request #99 from mlcommons/dev by @arjunsuresh in #100
• Merge pull request #99 from mlcommons/dev by @arjunsuresh in #101
• Fix format.yml by @arjunsuresh in #102
• Added typing_extensions deps to draw-graph-from-json-data by @arjunsuresh in #103
• Fixed the output parsing for docker container detect by @arjunsuresh in #104
• Improve setup.py by @arjunsuresh in #106
• Improve retinanet github action by @arjunsuresh in #107
• Fix retinanet github action by @arjunsuresh in #108
• Improve gh action by @arjunsuresh in #109
• Support GH_PAT for windows in push-mlperf-inference-results-to-github by @arjunsuresh in #110
• Merge from dev by @arjunsuresh in #105
• Code changes for supporting llama3_1-405b reference implementation by @anandhu-eng in #111
• Support hf_token in CM docker runs by @arjunsuresh in #114
• Fix github actions by @arjunsuresh in #115
• Update readme, inference submission cleanups by @arjunsuresh in #117
• Sync Dev by @arjunsuresh in #118
• Added Copyright by @anandhu-eng in #119
• Add copyright by @anandhu-eng in #121
• Inference submission generation improvements by @arjunsuresh in #120
• Update test-mlperf-inference-resnet50.yml by @arjunsuresh in #122
• Clean github action by @arjunsuresh in #123
• Sync <- Dev by @arjunsuresh in #124
• Sync Dev by @arjunsuresh in #126
• Fixes for MLPerf github action failures by @arjunsuresh in #127
• Merge changes for MLC by @arjunsuresh in #128
• Update test-mlperf-inference-abtf-poc.yml by @arjunsuresh in #129
• Update format.yml by @arjunsuresh in #133
• Fixes for MLC docker run by @arjunsuresh in #136
• Update check-broken-links.yml by @arjunsuresh in #137
• Fixes for ABTF docker run by @arjunsuresh in #138
• Fix PATH in dockerfile for ubuntu user by @arjunsuresh in #139
• Fix docker working with MLC by @arjunsuresh in #143
• Update test-mlperf-inference-resnet50.yml by @arjunsuresh in #144
• Update test-mlperf-inference-resnet50.yml by @arjunsuresh in #145
• Sync Dev by @arjunsuresh in #148
• Fixes for docker mounts by @arjunsuresh in #150
• Update check-broken-links.yml by @arjunsuresh in #151
• Fixes for nvidia-mlperf-inference by @arjunsuresh in #152
• Update module.py | Fix typo by @arjunsuresh in #153
• Sync Dev by @arjunsuresh in #154
• Fixes for nvidia mlperf inference by @arjunsuresh in #156
• Fix typo in docker_utils by @arjunsuresh in #157
• Cleanup by @arjunsuresh in #158
• Update test-nvidia-mlperf-inference-implementations.yml by @Arjunsure...