v0.10.3

v0.10.3

What's Changed

• moves deepspeed requirements into their own file; add deepspeed extras (backport #455) by @mergify in #546

Full Changelog: v0.10.2...v0.10.3

v0.11

What's Changed

• ci: Remove workflow that doesn't utilize training library (medium, -mp) by @booxter in #478
• Obey the FSDP sharding option default by @Maxusmusti in #486
• Change default internal sharding strategy to HYBRID_SHARD by @Maxusmusti in #488
• chore: Update the large e2e job to use fallback logic for selecting EC2 instances by @courtneypacheco in #491
• moves deepspeed requirements into their own file; add deepspeed extras by @JamesKunstle in #455
• chore: introduce dummy workflow by @cdoern in #497
• ci: Search for necessary instance for smoke job in multiple AZs by @booxter in #481
• ci: Fix -sdk fake workflow failure on actionlint by @booxter in #501
• build(deps): Bump actions/setup-python from 5.5.0 to 5.6.0 by @dependabot in #493
• use instructlab constraints-dev.txt in e2e test by @ktdreyer in #499
• build(deps): Bump step-security/harden-runner from 2.11.1 to 2.12.0 by @dependabot in #490
• ci: Use tox-current-env to reuse prepared venv with torch by @booxter in #482
• fix: extend nccl timeout by @cdoern in #507
• always log storage by @RobotSail in #510
• deps: Remove caps on ROCm dependencies by @courtneypacheco in #517
• ci: don't trigger pull_request_target job on its own workflow by @booxter in #519
• Enable pylint 'unused-argument' check by @fynnsu in #528

New Contributors

• @ktdreyer made their first contribution in #499
• @fynnsu made their first contribution in #528

Full Changelog: v0.10.0...v0.11

v0.10.2 - Remove ROCm dependency caps

What's Changed

• deps: Remove caps on ROCm dependencies (backport #517) by @mergify in #518

Full Changelog: v0.10.1...v0.10.2

v0.10.1 - Updating Default FSDP Sharding

What's Changed

• ci: Remove workflow that doesn't utilize training library (medium, -mp) by @booxter in #478
• Obey the FSDP sharding option default (backport #486) by @mergify in #487
• Change default internal sharding strategy to HYBRID_SHARD (backport #488) by @mergify in #489

Full Changelog: v0.10.0...v0.10.1

v0.10.0 - Updated FSDP Mixed Precision and Liger Kernel Model Option Support

What's Changed

• disables e2e-nvidia-l4-x1 test by @JamesKunstle in #454
• ci: Fix unit test run due to no tests found to execute by @booxter in #466
• ci: Don't run smoke tests when only irrelevant files are touched by @booxter in #460
• ci: don't waste ec2 resources on unit tests by @booxter in #464
• ci: Trigger unit test run on tox.ini change by @booxter in #469
• ci: Fix path filter for unit tests for the workflow file by @booxter in #461
• chore: Don't install pytest dependencies for coverage reports by @booxter in #468
• chore: Remove spell checks from the repo by @booxter in #458
• chore: Don't set ec2_runner_variant for unit tests by @booxter in #475
• Remove CHANGELOG.md by @booxter in #457
• Fix FSDP mixed precision setting and loss w/ accelerate by @Maxusmusti in #465
• fixes non-granite model instantiation with Liger Kernel by @JamesKunstle in #476
• ci: Install torch before flash-attn by @booxter in #474
• ci: Use pull_request as trigger for unit tests by @booxter in #473
• ci: Run unit tests for all supported python version, 3.11+ by @booxter in #472
• chore: Require python3.11+ by @booxter in #470
• chore: Drop pytest-asyncio by @booxter in #467
• chore: don't trigger unit tests for cuda and rocm requirements changes by @booxter in #463
• build(deps): Bump step-security/harden-runner from 2.10.4 to 2.11.1 by @dependabot in #452
• build(deps): Bump machulav/ec2-github-runner from 2.3.8 to 2.3.9 by @dependabot in #450
• build(deps): Bump aws-actions/configure-aws-credentials from 4.0.2 to 4.1.0 by @dependabot in #451

Full Changelog: v0.9.0...v0.10.0

v0.9.0

What's Changed

• build(deps): Bump machulav/ec2-github-runner from 2.3.8 to 2.3.9 by @dependabot in #431
• build(deps): Bump step-security/harden-runner from 2.11.0 to 2.11.1 by @dependabot in #439
• Adds Liger Kernels as optional optimization by @JamesKunstle in #441
• fix: model.forward now accepts return_dict via kwargs by @booxter in #443
• Adds smoke test workflow and tests by @JamesKunstle in #424
• change pytest targets. test-unit and test-smoke to unit and smoke by @JamesKunstle in #453

Full Changelog: v0.8.0...v0.9.0

Training Release v0.8.1

What's Changed

• fix: model.forward now accepts return_dict via kwargs (cherrypick) by @booxter in #446

Full Changelog: v0.8.0...v0.8.1

Training Release v0.8.0

What's Changed

• fixes unit ec2-name by @JamesKunstle in #409
• build(deps): Bump pypa/gh-action-pypi-publish from 1.12.3 to 1.12.4 by @dependabot in #410
• build(deps): Bump machulav/ec2-github-runner from 2.3.7 to 2.3.8 by @dependabot in #408
• syncs unit testing workflow setup with e2e setup by @JamesKunstle in #411
• ci: Add OpenAI keys into CI by @alimaredia in #415
• build(deps): Bump actions/setup-python from 5.3.0 to 5.4.0 by @dependabot in #418
• build(deps): Bump sarisia/actions-status-discord from 1.15.2 to 1.15.3 by @dependabot in #413
• ci: Don't require secrets in medium e2e test by @danmcp in #421
• Using AutoConfig to load model config file. by @abhi1092 in #416
• build(deps): Bump aws-actions/configure-aws-credentials from 4.0.2 to 4.0.3 by @dependabot in #417
• build(deps): Bump step-security/harden-runner from 2.10.4 to 2.11.0 by @dependabot in #426
• build(deps): Bump aws-actions/configure-aws-credentials from 4.0.3 to 4.1.0 by @dependabot in #425
• Generic notebook support by @Maxusmusti in #432
• build(deps): Bump actions/setup-python from 5.4.0 to 5.5.0 by @dependabot in #435
• Updating README and adding reasoning SFT example by @Maxusmusti in #436
• Updates data processing logic to remove dependency on hardcoded chat templates by @RobotSail in #428

New Contributors

• @abhi1092 made their first contribution in #416

Full Changelog: v0.7.0...v0.8.0

v0.7.0

What's Changed

• docs: include docs on installing deepspeed w/ cpuadam by @RobotSail in #333
• ci: Upload phase 1 & phase 2 training logs for loss graphs by @alimaredia in #356
• Add disk check after tests run by @danmcp in #361
• Updated token masking for new data "unmask" option (for pretraining samples) by @Maxusmusti in #357
• build(deps): Bump hynek/build-and-inspect-python-package from 2.10.0 to 2.11.0 by @dependabot in #366
• build(deps): Bump pypa/gh-action-pypi-publish from 1.12.2 to 1.12.3 by @dependabot in #364
• build(deps): Bump step-security/harden-runner from 2.10.1 to 2.10.2 by @dependabot in #355
• feat: add discord e2e status reporting by @RobotSail in #376
• Adjust to slack-github-action 2.0 api changes by @danmcp in #351
• build(deps): Bump slackapi/slack-github-action from 1.27.0 to 2.0.0 by @dependabot in #349
• adds pytest to tox via py3-unit by @JamesKunstle in #378
• build(deps): Bump rhysd/actionlint from 1.7.4 to 1.7.6 in /.github/workflows by @dependabot in #383
• build(deps): Bump DavidAnson/markdownlint-cli2-action from 18.0.0 to 19.0.0 by @dependabot in #381
• gh/actions unit test workflows by @JamesKunstle in #384
• changes Fast unit CI runner, m8g->m7i by @JamesKunstle in #389
• chore: Change default temporary write directory in all e2e CI jobs from tmpfs to /home/tmp by @courtneypacheco in #390
• feat: retain only last checkpoint directory by @leseb in #358
• fix: --keep_last_checkpoint_only does not accept any values by @courtneypacheco in #397
• build(deps): Bump rhysd/actionlint from 1.7.6 to 1.7.7 in /.github/workflows by @dependabot in #400
• build(deps): Bump hynek/build-and-inspect-python-package from 2.11.0 to 2.12.0 by @dependabot in #406
• build(deps): Bump actions/stale from 9.0.0 to 9.1.0 by @dependabot in #405
• build(deps): Bump sarisia/actions-status-discord from 1.15.1 to 1.15.2 by @dependabot in #403
• build(deps): Bump step-security/harden-runner from 2.10.2 to 2.10.4 by @dependabot in #402
• build(deps): Bump DavidAnson/markdownlint-cli2-action from 19.0.0 to 19.1.0 by @dependabot in #401
• Remove optimum dependency by @fabiendupont in #407

New Contributors

• @alimaredia made their first contribution in #356
• @courtneypacheco made their first contribution in #390
• @leseb made their first contribution in #358

Full Changelog: v0.6.1...v0.7.0

v0.6.1

What's Changed

• fix: disable loss exporting for medium training job by @RobotSail in #347
• build(deps): Bump DavidAnson/markdownlint-cli2-action from 17.0.0 to 18.0.0 by @dependabot in #348
• Update Dependencies to Move DeepSpeed to CUDA Extras by @Maxusmusti in #350

Full Changelog: v0.6.0...v0.6.1