0.26.5

Release Notes

0.26.5

Changelog

• cfa7730 chore: bump version: 0.26.5-rc3 -> 0.26.5
• 2755e5e docs: add release notes for 0.26.5 (#8564)
• 617fb0c chore: bump version: 0.26.5-rc2 -> 0.26.5-rc3
• 5eff3b4 chore: bump version: 0.26.5-rc1 -> 0.26.5-rc2
• 54aa7a8 feat: add the slots property to the props (#8498)
• 52adafb fix: allow slots per trial to be 0 [WEB-1871] (#8521)
• 6eb6bf0 fix(api): delete experiment error handling corrections (#8510)
• 54f2cec fix: trial spinner (#8528)
• 5cfc4cc chore: bump version: 0.26.5-rc0 -> 0.26.5-rc1
• 5f4a7f4 fix: ensure filter columns are valid when selecting special columns (#8517)
• 189b4c1 fix(agentrm): resource pools must filter agents.list() by name (#8509)
• f47bde5 ci: add date to python cache, fix moto linting issue (#8493)
• 22d6d25 chore: bump version: 0.26.5-dev0 -> 0.26.5-rc0
• 275ea84 chore: lock api state for backward compatibility check
• cea9ff4 fix: Experiment state now is an ExperimentState (#8457)
• 06b7b79 fix: tqdm logs within wrap_rank [MLG-1236] (#8488)
• f50f7db fix: use explicit e.state in bulk experiment delete query (#8491)
• 0f65698 fix(rm): tasks shouldn't hang on restore failures (#8486)
• f847b26 chore: Revert "test: quarantine GPU execution of test_task_logs (#8261)" (#8484)
• f085e10 fix: failing custom searcher due to ExtraEnvVars being overwritten (#8490)
• 88f64f6 chore: update libraries (#8463)
• 2e48a6f chore: migrate detaileduser and experimentitem types to io-ts (#8477)
• d8ed945 chore: add aliases to det dev commads (#8156)
• a2730cb feat: adding PACHD_ADDRESS and DEX_TOKEN to task env (#8473)
• 837bc29 chore: Update Hew Version to 0.6.12 (#8481)
• 1ee9b81 chore: clean up version dropdown update script (#8415)
• 398f879 docs: Add requirement and known issue for singularity-suid (#8478)
• 64e299e fix: wrong skip experiment config regex for log policies (#8475)
• 0b4e1d2 chore: cleanup some spurious cluster logs (#8468)
• 8c9dfbf fix: add delete cascade to generic metrics (#8469)
• 0b28148 ci: register unit pytest marks (#8470)
• 815f5ae fix: Kill task permission on interactive page (#8358)
• e67807d chore: preserve CI logs when bringing an AWS cluster down (#8461)
• 50d40bd chore: update trial complete or early exit to always notify searcher (#8466)
• 6bdf061 Update k8s install info (#8465)
• e06b472 chore: export AddUserTx (#8458)
• 16ea5a0 chore: introduce and use observables with improved update checking (#8405)
• 5805df2 chore: set up ownership for .circleci [skip ci] (#8402)
• c2a211d fix(api): handle delete experiment failures correctly (#8459)
• dfb4dc5 chore(actors): remove pkg/actor (#8452)
• 167e237 chore: add error check to KillNTSC (#8441)
• 4edbc7f chore: log RestoreAllCommands error (#8454)
• 9d71abd Fix minor issues including hard coded reference (#8427)
• 756a79c chore: bump CI node version to 20.9.0 (#8455)
• 3090e42 chore: use ResourcePool info for consistent capacity calculation [WEB-1796] (#8447)
• 192a2b3 chore(actors): remove pkg/actor usage from agentrm (#8395)
• 5ded38d chore: bump version: 0.26.4-dev0 -> 0.26.5-dev0
• 4339f67 docs: add release notes for 0.26.4 (#8451)
• a25e4f5 fix: add back pin icon in experiment list header (#8429)
• adb5191 ci: store npm log artifacts (#8449)
• e29006b fix: det slot task name for no-permissions RBAC users (#8416)
• 695a648 fix: SDK list_checkpoints not defaulting to searcher metric sort (#8448)
• 1ddad7d feat: add Topology into the RP details page (#8276)
• 2f7dda6 ci: cache install Python (#8426)
• cde18df fix: Calculate allocation bar stats same as overview [WEB-1822] (#8431)
• 9a48ff1 docs: Update upgrade instructions (#8346)
• e9a199a fix: k8s autoscaling nodes not counted towards RP (#8439)
• bd19e7a chore: command actor refactor & add intg test [DET-9660] (#8136)
• 99c4cea test: create a test for delete-tensorboards via det e delete (#8336)
• 7b7d1eb feat: Add remote user settings to Users table [WEB-1798] (#8397)
• 8051039 ci: fix linting with responses==0.24.1 (#8436)
• b087c10 chore: add version dropdown url for previous release (#8437)
• 8e9c505 test: fix model registry rbac wrong user regression (#8420)
• 292c75d fix: new experiment list tooltip styling (#8433)
• 1b47bf0 ci: delete broken fixture (#8428)
• bf07e61 fix: Wrap older modals in theme class [WEB-1824] (#8432)
• 0c8fad9 chore: filterformstore comment re: change tracking (#8386)
• 1041e56 fix: replace antd select with hew select (#8424)
• aa34aa7 feat: add workspace/project creation/deletion (#8430)
• 2e0a5a2 feat: client gets list_models, too. (#8425)
• 69df80e Revert "feat: Client gets list_models, too."
• d1343ca feat: Client gets list_models, too.
• a1e660d chore: converting SearchGroupsWithoutPersonalGroups into tx (#8419)
• 6ba688d test: fix TestAddAndRemoveBindings flake (#8423)
• d4c4195 chore: update Column and Row from Hew (#8412)
• d1d09e7 docs: Update non root container instructions (#8273)

0.26.4

Release Notes

0.26.4

Changelog

• bf665ae chore: bump version: 0.26.4-rc4 -> 0.26.4
• 2f86950 docs: add release notes for 0.26.4 (#8451)
• f2ef0fe chore: bump version: 0.26.4-rc3 -> 0.26.4-rc4
• f0a37a9 fix: Calculate allocation bar stats same as overview [WEB-1822] (#8431)
• 9acfbf2 chore: bump version: 0.26.4-rc2 -> 0.26.4-rc3
• 9dd0211 fix: k8s autoscaling nodes not counted towards RP (#8439)
• 3bf6647 chore: bump version: 0.26.4-rc1 -> 0.26.4-rc2
• 47397a4 fix: new experiment list tooltip styling (#8433)
• 680ac02 ci: fix linting with responses==0.24.1 (#8436)
• d4200d2 chore: add version dropdown url for previous release (#8437)
• 0a4b6bc test: fix model registry rbac wrong user regression (#8420)
• 242ff97 fix: Wrap older modals in theme class [WEB-1824] (#8432)
• e3c109a chore: bump version: 0.26.4-rc0 -> 0.26.4-rc1
• 22e18ae fix: replace antd select with hew select (#8424)
• 00d349a feat: add workspace/project creation/deletion (#8430)
• 9f727fe feat: client gets list_models, too. (#8425)
• b8c1be7 chore: update Column and Row from Hew (#8412)
• 4e6fd52 chore: bump version: 0.26.4-dev0 -> 0.26.4-rc0
• e9a457d chore: lock published urls to preserve redirects
• 2fae9ba chore: add docs dropdown link for new version
• 6c3bf84 chore: make insert-dropdown-url.sh executable (#8418)
• b5ca7f4 chore: fail deployment if launching part of the service fails (#8409)
• 8498674 fix: allow --json in det master config CLI command (#8413)
• d123932 fix: Place modal inside of ResourcePoolCard (#8414)
• ff19924 chore: Add eslint rule for ?? operator (#8410)
• d56b3ae chore: convert DOS line endings to Unix (#8411)
• c1219eb fix: Hide stats card when 0 on cluster page (#8359)
• da77efb fix: added permission check on GetAllocation (#8281)
• 3b0550c chore: Bumpenvs 0.26.4 (#8407)
• e48d03d fix: user flag to prompt for password during user requests (#8158)
• 513e6d7 fix: Project and Workspace cards wrap modal divs (#8378)
• 2497d84 chore: export AddUserTx (#8403)
• ad764f0 refactor: implement Glossary component from Hew (#8385)
• 52326d1 feat: change cli command for patch master log config DET[9720] (#8054)
• 1e9155d chore(type): stricter tsconfig (#8349)
• 16f18cc chore: revert task obfuscation lint failures (#8406)
• dde3156 chore: Implement Theming updates in Determined [WEB-1726] (#8388)
• 4edfc3c ci: move packaging test to test-e2e-longrunning (#8381)
• d3c208a ci: cache go modules deps and build cache (#8383)
• 8924996 chore: temporarily disable CI upload job (#8399)
• 356f651 Revert "chore: temporarily disable upload_test_results job step"
• 6dd9701 chore: temporarily disable upload_test_results job step
• ba49dbd ci: up parallelism for slowest test_e2e premerge tests (#8374)
• 5f3e556 ci: finish removing growforest (#8389)
• 62084e2 fix: NTSC task and slot viewing obscured for RBAC users with no Viewer Permissions (#8311)
• 0254f7d chore: fix nil ptr on allocation.Proto() (#8372)
• 119e759 chore: fix profiler test in CI (#8382)
• b428d5e feat: add hide column header menu item to explist (#8342)
• 7ae0501 chore: update the lore service port (#8375)
• 052cf8d feat: Cluster historical usage charts move to UI Kit LineChart [WEB-1786] [WEB-1764] (#8327)
• 819948d feat: clear filter from experiment table header (#8376)
• a590999 test: fix slow delete_checkpoint test (#8377)
• b0505db chore: Job/task displays Running instead of Scheduled (#8335)
• 1d64941 chore: short dsat e2e tests (#8288)
• 6afa836 chore: fix CI mnist_pytorch (#8364)
• 4d3eaab chore: Update Horovod Cycle Time (#8362)
• d3b01cb docs: Add det pach tutorial (#8082)
• 7cebc30 fix: adjust card size on workspaces page (#8370)
• 5c93cb0 chore: enable more Go linters (#8333)
• a279967 fix: aws deployment can deploy priority scheduler (#8345)
• 3d9293c fix: fixed bug in error handling in experiment.go (#8339)
• 194bfd5 fix: Cell can be undefined in experiment list table (#8360)
• 1da92aa chore: bump environment images to ubuntu 18.04 [MLG-1194] (#8356)
• 990c56f chore: add list_experiments to experimental.client (#8361)
• 3a7d9ea fix(tests): lower e2e_gpu_quarantine parallelism (#8363)
• 4c48458 fix: patched remote users were able to login with password (#8337)
• baf5c96 chore: port over PyTorch example to use Trainer API [MLG-1181] (#8292)
• 235bd8f feat: delete TB files from the SDK (#8329)
• 2fe3d99 chore: update Typography from UI kit (#8323)
• 2b23674 fix: prevent carriage return in env from crashing deepspeed launcher (#8321)
• 461c307 chore: Remove DesignKit since it's now maintained in Hew [WEB-1790] (#8338)
• 5ee87ec fix: Set group name and number columns to handle Safari [DET-9948] [DET-9949] (#8355)
• 10deef9 fix(experiments): transient errors shouldn't leave trial hung (#8352)
• 512b9f3 chore: remove accidental mock commit (#8354)
• 9d17dbf feat: Show "-" for null values in data cells for experiment list (#8343)
• ea50987 fix: properly interpret flag values (#8326)
• 8b6fc68 fix: Allow SAML and OIDC logins to work differently [WEB-1797] (#8308)
• 274288e docs: fix linting failure (#8351)
• 73bf0e8 docs: log policies (#8302)
• 8418029 chore: ft slot capacity check for each trial [DET-9897] (#8213)
• 494ca57 fix: replace TODO with ctx for deleteTensorboard (#8332)
• cfde2f6 docs: Docs Version Dropdown Automation (#8340)
• 8e69941 chore: Remove examples/legacy (#8153)
• af995ba fix: cli is not a library! (#7891)
• bf0a03d test: fix ray.air.session import. (#8344)
• 9bb10cc ci: mypy fix for responses>=0.24.0 (#8341)
• b924b25 fix: add pin icon in dropdown (#8324)
• 62b7f3b chore: remove fit-content from TimeAgoc (#8328)
• 86d6962 chore: update determined-ui to hew (#8334)
• f580385 fix: metric group charts have more than one color (#8304)
• 1966373 feat: Add tensorboard delete command to CLI (#8227)
• 656c8b2 chore: bump version: 0.26.3-dev0 -> 0.26.4-dev0
• af43248 docs: add release notes for 0.26.3 (#8322)
• b262a3d chore: Update lore.yaml to use the new version
• d64a0ac chore: use a single .golangci.yml file (#8320)
• ad94d20 chore: Add progress bar from UI Kit [WEB-1675] (#8181)
• b3b5be0 feat: implement CodeSample from UI Kit [WEB-1677] (#8270)
• d723b7f docs: fix typo in user edit release note (#8319)
• 6e5d840 chore: initial experiment actor refactor (#8229)
• 8a1ff58 chore: use a single root level go mod (#8285)
• 3511abf chore: delete dead code (#8313)
• d0e6375 chore: add a new deployment type for aws (#8279)
• 3929e8c chore(actors): remove ctx usage in agent_state.go (#8267)
• 5bf1b87 ci: delete broken wait_for helper (#8312)
• 50535f1 test: quarantine GPU execution of test_task_logs (#8261)
• d5b8e80 chore: deployment's --dry-run option doesn't print template (#8303)
• ac89d44 fix: allow experiments with directory checkpoint storage to parse (#8310)
• 306c0c3 fix: Project info not presists when forking (#8307)
• dc1b131 chore: sort out issues after bringing EE e2e_tests into OSS (#8084)
• d182abe chore: slurm support for blocklist (#1111)
• efdf62b fix: return correct location URL for /Users SCIM API endpoint (#1115)
• 37a84d1 fix: ruamel.yaml fixes for EE
• 0ce925a chore: Update nightly tests that use legacy cifar10_pytorch (#1102)
• 6cad296 fix: update for error message change in product (#1098)
• 1e302b5 chore: update e2e tests affected by examples_pruning (#1100)
• ad3dcda chore: cleanup model registry rbac test
• a3ffb5d test: enable command run tests for PBS (#1073)
• 9dd0e42 test: enable command and deepspeed tests run on slurm/pbs (#1044)
• ea4f4c4 chore(templates): ee fixes for template rbac
• c48e48d fix: Test test_slurm_verify_home fails with podman and it shouldn't [FE-136] (#1028)
• 760a738 test: Add pytorch2 distributed e2e tests on slurm [FE-168] (#1007)
• b5aee79 chore: use longer running no op experiment when seeding workspace (#994)
• facbda9 test: run test_hpc_job_pending_reason only on gcp vm (#996)
• 393d0b5 ci: FE-133 Configure non agent slurm/pbs tests to skip without explicitly listing test names in circleci. (#977)
• fd15535 ci: add ee-only files to the import-restrictions linter exclusions.
• 9cf2a26 test: slurm/pbs test for pending reason (FE-90) (#960)
• b3c2ca3 chore(actors): allocation.go, ee side
• eb7d1a1 test: [ALLGCP] Add e2e test for HPC that verifies that user HOME is preserved (#972)
• f3a8b0e test: fix test_slurm.py lint error (#949)
• 71896f3 chore: FE-91: Update base images (slurm/pbs) to include a populated singularity_image_cache (#943)
• 5891567 feat: add rbac to api/v1/master/config [DET-9633] (#931)
• 0a5c32e ci: FE-72: Add test-e2e-pbs-*-gcp tests (#941)
• 4c233c0 feat: add rbac for strict job queue control (#927)
• 6e23aa2 chore: removed admin dependency from delete model/version (#912)
• 7c6c59e feat: rbac for templates (#909)
• e89cc08 ci: DET 9622: (ee) test_slurm.py::test_cifar10_pytorch_distributed failures (#919)
• c6ee094 fix: test_rbac goes to wrong url (#918)
• 6b34e0d fix: DET-9483 successfully run e2e_slurm_preemption tests as part of nightly workflow (#903)
• 4f6277d ci: FE-14 Migrate test-e2e-slurm to GCP slurmcluster (#879)
• f4507f7 tests: fix a miss indentation leading to missing project err (#878)
• 5cad9bb chore: fix a missing check for global permissions in jq (#874)
• bca3848 feat: add rbac support for reading job queue (#871)
• 5c79474 chore: update how we wait for tasks to be ready (#863)
• b292862 test: fix test_master_host [DET-9482]. (#851)
• 5375a08 ci: quarantine flaky slurm tests (#850)
• 8e44c6c fix: Patch groups test [DET-9473] (#845)
• 49d2e08 fix: fix bug with launching tensorboards on trials (#842)
• d4dcbe5 test: Fix and add e2e_slurm_preemption tests to nightly workflow [D...

0.26.3

Release Notes

0.26.3

Changelog

• bd74446 chore: bump version: 0.26.3-rc3 -> 0.26.3
• 162de31 docs: add release notes for 0.26.3 (#8322)
• a472745 chore: bump version: 0.26.3-rc2 -> 0.26.3-rc3
• bab1dad fix: allow experiments with directory checkpoint storage to parse (#8310)
• ec438f2 fix: adjust width size in group table (#8309)
• 0d49b53 chore: fix job service panic when workspace does not exist (#8306)
• 13525e8 fix: check externalConfig is enabled before setting det_jwt as auth header (#8298)
• 101f279 fix: undefined handling in CreateGroupModal (#8301)
• b9e64ab docs: quick fix for version dropdown (#8300)
• 48e915f chore: bump version: 0.26.3-rc1 -> 0.26.3-rc2
• 3c2cb53 ci: update wrapper config to always run
• 55c8a3a chore: bump version: 0.26.3-rc0 -> 0.26.3-rc1
• c138065 chore: bump version: 0.26.3-dev0 -> 0.26.3-rc0
• 862b41e chore: bump version: 0.26.2-dev0 -> 0.26.3-dev0
• b2e4b02 docs: add release notes for 0.26.2 (#8245)
• 989a0e3 fix: update bumpversion cfg for new CircleCI config (#8293)
• 54a12b8 fix: fix docs linting (#8291)
• 108cca7 docs: add documentation for Keras and PyTorch profilers [MLG-1094] (#8253)
• a7dfd47 chore: lock published urls to preserve redirects
• 8ae316c chore: lock api state for backward compatibility check
• d0c8273 chore: refactor filterformstore (#8239)
• 2992b82 feat: Updated LineChart in UI Kit [WEB-1700] (#8105)
• c6c3235 fix: login error message (#8240)
• 949ecf9 feat: support PyTorch Profiler in DeepSpeed trials [MLG-1095] (#8251)
• b2ea839 fix: use proper default cpu env image in the helm chart. (#8287)
• ccc78dd chore: install request as dev dep to fix proxy.js (#8286)
• af40f2c fix: only train for one batch in PyTorch Trainer test mode (#8260)
• cbab7e3 fix: dsat with all yaml formats (#8284)
• d17a2cc fix: migrate CLI from deprecated SDK methods (#8282)
• f43acd8 feat: directory checkpoint storage [DET-9594] (#8255)
• f375fbb feat: log policies (#8145)
• 862951f chore(actors): remove slots, slot proxy hacks from agentrm (#8266)
• 8d79c17 feat: webhook type task logs (#8175)
• b086982 test: fix delete experiments potential flake (#8283)
• fe06a0a fix: update copy for Agent UID/GID modal in the user mgmt UI. (#8278)
• 1fbb9f0 chore: restore set resource pool using job service (#8280)
• 0bea20b chore(actors): remove actors from resource aggregation (#8265)
• 76b4bfe feat: update add members to group (#8262)
• f24748c docs: Fix a 404 error (#8277)
• e4701b2 fix: remove Topology from the ResourcePoolDetail page (#8274)
• bec5edd chore(actors): refactor k8s rm without actors (#8264) [DET-9658]
• 46b9360 fix: icon size in TaskBar (#8263)
• 5c12463 test: don't commit Go mocks (#8258)
• a33d2b8 chore: ci should always setup python venv for caching (#8257)
• c09c529 ci: wart removal (#8147)
• 65219ca docs: fix inaccuracies in bind_mounts docs. (#8254)
• d3bd2cb feat: unify new/edit group modals [WEB-1741] (#8236)
• 3179505 build(deps): bump actions/setup-node from 3 to 4 (#8230)
• e0a7efc fix: support ruamel.yaml>=0.18.0 (#8237)
• 2365df6 docs: check for dropped urls in PRs (#8247)
• f9c6402 fix: deepspeed e2e_tests fail when environment variable contains a newline character [FE-256] (#8154)
• ec7e004 fix: prevent passing login r= param to relayState (#8244)
• ab4cf83 feat: Update "Add Members" to Workspace experience (#8195)
• 25adf7d fix: dont suggest registering or deleting a checkpoint that is already deleted (#8246)
• c52ef71 feat: add new CLI command to edit multiple fields at once (#8075)
• 1d86343 chore: move files from /Clusters into /Cluster [WEB-1730] (#8231)
• 15ee78f docs: Add article for using detached mode (#8217)
• 115d7c2 fix: GetTasks doesn't respect rbac (#8233)
• 47b37eb chore: update Avatar component in UI kit [WEB-1055, WEB-1734] (#8178)
• 4c59393 chore: indirect import for jobs to avoid import cycle (#8238)
• aa77386 chore: add back missing commit in Python SDK (#8206)
• 9a80d99 ci: alternate mechanism for running nightly tests (#8221)
• 27a279b refactor: use Nameplate component in NavigationSidebar [WEB-1057] (#8152)
• c0e3062 revert: fix: support ruamel.yaml==0.18.0 (#8235)
• ae839a8 fix: support ruamel.yaml==0.18.0 (#8228)
• bb7020a fix: inaccurate task queue time [DET-9912] (#8225)
• 12e23b0 fix: pin ruamel.yaml<0.18.0 (#8232)
• 164b920 feat: Update users and groups tab to reflect count (#8224)
• 2ac9661 chore: shorten distributed-quarantine name (#8234)
• c964987 docs: Apply minor edits (#8215)
• f61d250 chore: support redirect on auth failure for echo routes (#8196)
• f251add docs: update the status of TLS security for notebooks. (#8191)
• 54551c5 refactor: CliError does not need e_stack property. (#8179)
• 7c6b2d0 chore: deprecate apex support. (#7526)
• bfda78d docs: quick fix for version dropdown (#8223)
• 5929161 docs: Fix broken Slack links (#8220)
• b076179 chore: refactor actor system out of internal/job (#8174)
• facdc30 fix: remove 'contents' parameter from remove_notes (#8209)
• 448162d docs: Edit setup checklist (#8214)
• 164579f feat: update group table (#8194)
• 8f5e883 fix: Multi-trial visualizations switch from metricType to group string [DET-9896] (#8137)
• 834bd2f chore: remove WebSocket actor (#7552)
• b6e28d7 fix: prevent extra updates when observing settings store (#8212)
• 707f77d fix: Trigger function updates new user modified_at without error (#8210)
• f9c8a86 chore(actors): refactor k8s' resource_pool.go without actors (#8186) [DET-9657]
• ea3d6bf docs: sort articles by weight custom extension (#8208)
• 464fb54 docs: Create new advanced setup section (#8203)
• dd1c6f0 docs: toctree tile css (#8207)
• 53b9bdb fix: CLI uses default where pagination not included in args [DET-9908] (#8192)
• 079304a fix: filter agents/nodes by poolName (#8205)
• 3af965f chore: move ui kit to separate repo (#8104)
• 8735acf docs: remove references to PyTorchTrialContext.from_config() (#8187)
• 03383f2 fix: fix issue with db migrations (#8193)
• 2ff18bc fix: Display byte axis values using humanReadableBytes [DET-9906] (#8189)
• bd5e628 feat: add "Topology" section to the cluster UI (#8108)
• 6dfde99 fix: icons should appear in safari (#8190)
• 36ad3ca chore(actors): refactor pods.go without actors (#8170) [DET-9901]

0.26.2

Release Notes

0.26.2

Changelog

• 85b5135 chore: bump version: 0.26.2-rc4 -> 0.26.2
• 25e578d docs: add release notes for 0.26.2 (#8245)
• 1f7945e chore: bump version: 0.26.2-rc3 -> 0.26.2-rc4
• 87c11fb fix: inaccurate task queue time [DET-9912] (#8225)
• 07ea3b2 fix: pin ruamel.yaml<0.18.0 (#8232)
• 6e8a762 docs: quick fix for version dropdown (#8223)
• 8896d23 fix: remove 'contents' parameter from remove_notes (#8209)
• b3fdf9e chore: bump version: 0.26.2-rc2 -> 0.26.2-rc3
• 5585f38 fix: prevent extra updates when observing settings store (#8212)
• 826f633 chore: bump version: 0.26.2-rc1 -> 0.26.2-rc2
• 829c3bf fix: Trigger function updates new user modified_at without error (#8210)
• 0b0d268 fix: CLI uses default where pagination not included in args [DET-9908] (#8192)
• b0071c2 chore: bump version: 0.26.2-rc0 -> 0.26.2-rc1
• f049bd6 fix: fix issue with db migrations (#8193)
• 1f4bbe2 fix: icons should appear in safari (#8190)
• 5d3a80d chore: bump version: 0.26.2-dev0 -> 0.26.2-rc0
• 57fee58 chore: lock published urls to preserve redirects
• 883135a chore: various deprecations to standardize SDK get/list/iter (#8165)
• 0ab908e chore: add release notes for Python SDK (#8184)
• 2c09e10 docs: fix redirects (#8188)
• bfd20f1 docs: update experiment config reference for records_per_epoch in PyTorchTrials (#8185)
• 1fb50cd fix: pin icon should adapt to theme colors (#8183)
• aff81a5 chore: migrate SDK to a generic OrderBy [MLG-1056] (#8171)
• 88549ee Revert "feat: use max_results in SDK's list_trials (#8173)" (#8182)
• c29d086 fix: tensorboard sync for profiler data, Core API v2 managed mode [MLG-1063] (#8163)
• 8324bf2 fix: Change oicd client secret env var name to comply with naming convention (#8113)
• 3946185 docs: Restore path to singularity file (#8180)
• d12373f chore: reinstate core_api example e2e tests (#8148)
• 16ef0bb feat: use max_results in SDK's list_trials (#8173)
• ee2e633 refactor: add delete cascade to tables affected by experiment deletion (#8016)
• 404831b chore: final trial actor refactor (#8164)
• eb98aeb fix: case insensitive member search (#8166)
• b76dd90 fix: Learning curve point click (#8155)
• 95f2ad7 chore: task resources actor refactor (#8157)
• 1bff62e chore: configure a training port offset (#8125)
• e4d5ad1 docs: Modify the info architecture (#8100)
• 556022a chore: change CSS values (#8151)
• 88d9533 fix: Theme dropped after page refreshing (#8139)
• 515f128 refactor: update UI Kit Icon component [WEB-1699] (#8122)
• 82b4d31 chore: Post pruning hotfixes (#8141)
• 67f6dc1 Python SDK v1 (#8005)
• 3a24611 docs: fix python-sdk reference syntax (#8146)
• e7247a5 docs: Restore core api integer incrementing tut (#8115)
• 2a7f141 chore: add option to proxy requests to internal service (#8044)
• d40dee3 fix: Move setting of playwright browsers path (#8144)
• 31c2515 chore: update metrics documentation (#8118)
• b973356 chore: Add Message component to UI kit [WEB-1056] (#8133)
• 5df3d57 chore: remove actors from resource manager interface (#8126)
• f9a61b4 chore: upgrade to node v20 and npm audit [WEB-1662] (#8036)
• 38ff7b5 docs: update link in prometheus docs (#8138)
• ea8d760 chore: Examples pruning (#8140)
• 8b9ec03 fix: move couldn't-connect-to-master message into ship_logs.py (#8127)
• f52f7c6 docs: Apply style guide edits (#8134)
• 2f65eec chore: bump version: 0.26.1-dev0 -> 0.26.2-dev0
• ee3c478 docs: add release notes for 0.26.1 (#8131)
• 5d55da1 ci: dannys/reenable docs autoassign (#8012)
• 017e0f2 docs: Remove ref to fluent bit (#8128)
• 42535c2 chore: add hf context to all dist e2e [DET-9893] (#8119)
• 7e6689b chore: make agent.yaml readable by default (#8095)
• 8d2b5ce fix: hotfixes for ship_logs.py (#8116)
• 233cd95 feat: add allocation exit status to db and implementation for get allocation (#7897)
• 22f1f92 chore: remove task allocation group actor ref (#7853)
• 42bfb72 fix: Return ResourcePools with a fixed order (#8103)
• 3b42b91 chore: write log shipper in python [MLG-993] (#7974)
• ae7c415 fix: DET_CERT_MASTER_FILE=noverify det shell open (#8110)
• 3e1b83e docs: redirects.py moves subdirectories properly (#8111)
• 29aa772 fix: fix primary key for allocation_accelerators table (#8106)
• f9e9a45 build: print sphinx-build command on make build (#8109)
• a51ef7c fix: update WorkspaceMemberAddModal when new user or group is created [WEB-1111] (#8069)
• ff9b77a fix: trigger autoupdate_users_modified_at by username change (#8093)
• 4899df7 feat: Edit experiment from list (#8086)
• d638303 feat: Dont allow users to add deactivated users to a workspace (#8073)
• f559eee fix: single point different axis ranges (#8096)
• 0c54010 fix: Empty / NotEmpty operators for descriptions, tags [WEB-1751] (#8090)
• 469c2ae perf: improve task stats IMAGEPULL performance (#8067)
• e7751fe chore: auth task logs [DET-7554] (#8089)
• 588d817 chore: cleanup allocation exit logic (#8088)
• e2b51e7 feat: add implementation for get and set acceleration data api and intg test [DET-9748] (#7856)
• 5483d12 fix: ignore last_auth_at to update modified_at on users table (#8091)
• 0f8c1d4 fix: Trial data loading state (#8083)
• 6a91c3d fix: Metric type is blank in comparison chart (#8085)
• 1819810 chore: allow attaching select dropdown to select container (#7940)
• e5b2e35 refactor: removed reference removed agents/./slots/. endpoint (#7424)
• ecc0f6b chore: support "formData" in swagger bindings parsing (#8078)
• 9dfc0a3 feat: Hide deactivated user in ws members list (#8077)
• 89c8da2 feat: introduce batch actions in user management page (#8056)
• ca63335 docs(performance): add sections to help with getting started (#8079)
• 9ea05a1 chore: change master config to k8s secret (#8053)
• 58fab08 fix: Experiment name settable in fork config (#8081)
• 6006d42 fix: Show experiment loading state (#8066)
• 8073a2d fix: order of API paths in proto for AssignMultipleGroups (#8080)
• 5be764b feat(performance): implement slack report send [INFENG-234] (#8045)
• 18d7286 fix: shell open gives unfriendly message on terminated (#8074)
• fc2e8bf test: fix test_pytorch_parallel logs check (#8064)
• ca763a4 fix: det deploy aws list results were incomplete. (#8062)
• c8edfab chore: cleanup pod informer logging (#8072)
• 14bd162 docs: quick fix for version dropdown (#8070)
• bcfdeac chore(tests): workaround data races in uptrace/bun (#8065)
• 5d56e95 chore(tests): fix data races in webhook integrations (#8061)
• ce5bcf2 ci: backend owns e2e_tests folders cluster, command, and template (#8037)
• 3075820 chore(tests): fix data races in streaming API intg tests (#8059)

0.26.1

Release Notes

0.26.1

Changelog

• a6b26b0 chore: bump version: 0.26.1-rc3 -> 0.26.1
• 4bd3dcb docs: add release notes for 0.26.1 (#8131)
• de1526b chore: bump version: 0.26.1-rc2 -> 0.26.1-rc3
• 6e19285 fix: Return ResourcePools with a fixed order (#8103)
• 355cb62 chore: bump version: 0.26.1-rc1 -> 0.26.1-rc2
• 3f97397 fix: Trial data loading state (#8083)
• 740e730 fix: trigger autoupdate_users_modified_at by username change (#8093)
• 288db9f fix: single point different axis ranges (#8096)
• d972837 fix: Empty / NotEmpty operators for descriptions, tags [WEB-1751] (#8090)
• 3c5b281 fix: ignore last_auth_at to update modified_at on users table (#8091)
• b086b84 chore: bump version: 0.26.1-rc0 -> 0.26.1-rc1
• 2283723 fix: Metric type is blank in comparison chart (#8085)
• 90dfb54 fix: Experiment name settable in fork config (#8081)
• ce7549a fix: shell open gives unfriendly message on terminated (#8074)
• 663f4d8 docs: quick fix for version dropdown (#8070)
• 180d1b3 chore: bump version: 0.26.1-dev0 -> 0.26.1-rc0
• 1a96b8a chore: lock published urls to preserve redirects
• d1c3fa4 Revert bump version (#8063)
• 9b3b65a chore(tests): fix data race in grpclog init during integrations (#8058)
• ef0ef50 chore: lock published urls to preserve redirects
• b241b78 chore: bump version: 0.26.1-dev0 -> 0.27.0-dev0
• b14cc40 test: stable diffusion example tests [MLG-903] (#7855)
• 4cb3151 docs: Improve upgrade instructions (#8032)
• aa736a2 fix: detached mode tensorboard storage support [MLG-872] (#7992)
• d11f6c9 fix: fake cert gen.sh generated broken certs (#8055)
• a8a3399 fix: bring back logging for core.train.report_* calls. (#7975)
• f4524ac fix: pass final forked config to server for new experiment (#8051)
• 5ff6e9c feat: update InlineForm inputs (#8033)
• 9c05b00 feat: updating agent user group affects users.modified_at (#8052)
• 9414539 fix: fasterrcnn image not found [MLG-516] (#8047)
• 6aca426 chore(tests): fix data races in k8s tests (#8049)
• 764f407 chore(tests): fix data races in telemetry tests (#8048)
• 90fb051 chore: rename 'cancel' to 'stop' for experiments and trials [WEB-291] (#8038)
• a4c244e fix: command resolve pool defaulting to workspace 0 instead of 1 (#8050)
• ccefc6d chore: allow passing clusterID in master config (#8042)
• 1214a2b fix: Return ResourcePools rather than names (#7990)
• 7aa8541 fix: govcloud agent AMIs are out of sync with bumpenvs [MLG-986] (#7983)
• be1f734 chore: add DatePicker to UI Kit [WEB-1674] (#8040)
• 5bd7b8a feat: SDK can list workspaces. (#7765)
• c0467b0 ci(performance): create initial gha workflow [INFENG-224] (#7969)
• 9dcedad fix: k8s custom pod spec affinity would get ignored (#8043)
• f4d3c47 fix: Note UI respects project permissions (#8028)
• 338d2f6 fix: rename last_login to last_auth_at (#8022)
• 34373ac feat: Batch actions for multiple users into one request [WEB-1640] (#7971)
• 1d85304 test: fix shell open test flake (#8035)
• 827d9a6 feat: Filter user list by role id for EE (#7988)
• e57b8d6 fix: Hide checkpoint deletion btn when already deleted (#8039)
• 04609e0 fix: Remove the unexposed GetJobQStats from RM interface and all RMs (#8030)
• d00ddc9 chore: Add error case and tests to Loadable [WEB-1333][WEB-1711] (#8025)
• 78dfc83 fix: Support longer titles on HParam scatter plots (#8031)
• fd9b406 fix: nil ptr for Proto() on users who haven't logged in (#8029)
• 846fbc3 fix: add css formating for miltiple input errors (#8011)
• 42a4911 chore: adding an example for distributed batch inference for mnist (#7976)
• c83dfe6 docs: fair-share scheduling policy [skip ci] (#7981)
• ccfda89 fix: Add fetch to resource pool bindings page (#8023)
• 4b085ff chore: move Loadable to kit [WEB-1688] (#7973)
• 67fdacb fix: sort files for conflict resolution in sharded checkpoints (#8014)
• 10fcb10 feat: add tooltip linebreak to the project card (#7995)
• 4b4ad8d fix: add existing "non-setting" query parameter into the settingsToQuery function (#8013)
• f021ac2 test: fix master intg user flake (#8019)
• 8937f2b fix: align default markdown font-family to theme [WEB-617] (#8009)
• e03e469 feat: "det notebook|shell|tensorboard open" doesn't error when task not ready (#8008)
• 109e69f fix: tensorboard deletion when det e delete [DET-9844] (#7997)
• e3ffc4f docs: Fix minor issues (#7999)
• 2e77f52 fix: regular integer spacing of chart ticks [WEB-1714] (#8010)
• f1ae472 feat: add last seen column in user management table (#7991)
• d2a5505 refactor: remove external dependencies from UI kit [WEB-1689] (#7968)
• e86abd5 fix: fix sorting in GetUsers endpoint (#8001)
• 69921eb feat: commands download user files at startup so k8s can support larger context directories [DET-8830] (#7889)
• 86e6f47 fix: report searcher progress according to reporting period (#8006)
• 5b2f238 ci: more precisely select files for splitting in E2E tests (#7989)
• f918188 fix: show deep files in experiment code viewer (#7945)
• 1d2b60b fix: data fetch shouldn't interrupt editing model version description [WEB-1703] (#8000)
• 7dff1f2 chore: bun debug mode off (#7996)
• 0241795 fix: Version dropdown in docs is scrollable (#7994)
• 46da826 ci: disable docs review action [skip ci] (#7982)
• 390e0ac chore: add tests for postgres_users.go (#7875)
• a60c0d7 fix: Hide action menu on dashboard project cards (#7986)
• a2fb7ef chore: bump version: 0.26.0-dev0 -> 0.26.1-dev0
• 7eaf361 docs: add release notes for 0.26.0 (#7987)
• 25439fc chore: put grpc panics and other logs into 'master logs' (#7965)
• 8c373cf fix: Force GCP node name length to be less than maximum length (#7964)
• 7f21b4f chore(templates): refactor templates to their own package (#7876)
• b9fb3eb chore: Add toast to UI kit (#7950)
• d45e22b feat: Support filter by status and role for users (#7953)
• 5b5fb2f chore: fix det deploy aws requiring --db-size (#7984)
• e254bf3 docs: Update custom pod specs page (#7970)
• 3d1ec1f fix: Update Tasks Stats Causes Deadlock [DET-9853] (#7980)
• 5c5e0ec feat: single experiment continue [DET-9703] (#7764)
• 34c5bdb docs: document prometheus auth (#7957)
• 6e3cad3 chore(codeowners): map performance dir to web team (#7916)
• 9a5cfd7 fix: handle nil actor message and nil actor errors in agent RM (#7951)
• d25612b feat: Add instance flavor and size arguments for det deploy aws [INFENG-227] (#7931)
• e031a2f fix: update modified_at by insert in user table (#7949)
• 2d65e82 fix: Change measure of text lines in log containers [WEB-1664] (#7860)
• 14a779c chore: add last_login column to users table/model (#7948)
• f4e3638 chore: fix docs reference in cli [MLG-891] (#7926)
• a4fdd0a chore: Improve failure diagnostics in shell test [FE-216] (#7932)
• b576b6b chore: less verbose mockery output (#7822)
• 9df980c chore(performance): add initial Makefile and README (#7914)
• 47a4070 chore: handle case where steps completed is more than max length (#7816)
• 1193acd chore: log k8s nil event objects at trace level and ignore (#7962)
• 1c86762 chore: max_slots_per_pod can be per resource pool [DET-9771] (#7923)
• 3267b1f fix: return searcher_metric_value as-is (#7961)
• b5845b7 fix: singularity agent env variable (#7960)
• 19d703b fix: mitigate user settings race conditions (#7905)
• f4382a7 fix(db): handle erroneous nulls from the summary metric migration (#7958)
• 65cabae fix(experiments): don't transition experiment to "" state on crash (#7956)
• aa30e86 docs: quick fix for version dropdown (#7952)
• 18e2f1f fix(scheduler): tolerate missing groups in priority scheduling by skipping them (#7947)
• 4b19928 chore: bunify & tidy up internal/user (#7886)
• 3f067a3 fix(allocation): allocation lifetimes should contain resource lifetimes (#7944)
• b6b5a84 Remove references to --auto-bind-mount (#7910)
• 2ba2580 chore(deps): bump tibdex/github-app-token from 2.0.0 to 2.1.0 (#7938)
• 2baaf30 fix: clear selection after action (#7921)
• 4db1c08 refactor: css in docs (#7934)
• a392dc8 fix: button in 404 page (#7936)
• 0e3f4d0 fix: avoid hiding tabs in single trial experiment [WEB-1651] (#7941)

0.26.0

Release Notes

0.26.0

Changelog

• 29705a8 chore: bump version: 0.26.0-rc3 -> 0.26.0
• 084e485 docs: add release notes for 0.26.0 (#7987)
• 2882e78 chore: bump version: 0.26.0-rc2 -> 0.26.0-rc3
• 623774e fix: Update Tasks Stats Causes Deadlock [DET-9853] (#7980)
• c11c5e4 fix: handle nil actor message and nil actor errors in agent RM (#7951)
• df6c317 fix: update modified_at by insert in user table (#7949)
• 9a795c1 chore: bump version: 0.26.0-rc1 -> 0.26.0-rc2
• d330154 chore: log k8s nil event objects at trace level and ignore (#7962)
• a8465d8 fix(db): handle erroneous nulls from the summary metric migration (#7958)
• ab9d933 fix(experiments): don't transition experiment to "" state on crash (#7956)
• 7b506f9 fix(allocation): allocation lifetimes should contain resource lifetimes (#7944)
• 2b12176 chore: bump version: 0.26.0-rc0 -> 0.26.0-rc1
• 78482b2 docs: quick fix for version dropdown (#7952)
• 0893e30 fix: clear selection after action (#7921)
• a9a382f fix: button in 404 page (#7936)
• 7f85454 chore: bump version: 0.26.0-dev0 -> 0.26.0-rc0
• 9ac8f82 chore: lock published urls to preserve redirects
• 6ba27fe chore: lock api state for backward compatibility check
• b52b3a6 chore: bump version: 0.25.2-dev0 -> 0.26.0-dev0
• c2cea7d chore: include api op and param description in py bindings (#7798)
• c98cc07 feat: Allow passing in swagger json as an argument (#7843)
• 482285f docs: Add another top nav link (#7933)
• d9e1bb5 chore: track dead code [WEB-258] (#7924)
• 537cc3d docs: Update launcher version to 3.3.8 for consistency with docs (#7915)
• 5b859c6 fix(cli): det model describe should call GET /model not GET /models (#7912)
• 262c33a docs: Clarify weighted fair-share scheduling policy (#7913)
• 4a486bb feat: Add performance tests for endpoints used in the WebUI initial load [WEB-1459] (#7906)
• cc360ac feat: Add workspaces to the SDK client (#7883)
• c87ca94 fix: api_command.go does not merge map values when overrides TaskContainerDefaults [FE-114] (#7887)
• dc40688 chore: update docs ownership per discussion [INFENG-225] [skip ci] (#7907)
• ee79213 test: fix e2e_tests ray dependency. (#7925)
• 688ff63 fix: align items in task list (#7894)
• 694f44a feat: submit forms in modals by pressing enter [WEB-1130] (#7857)
• 376ea50 fix: Display data point in line chart when epoch is 0 (#7898)
• 8eff8ac chore: update user docs (#7902)
• 4441ac8 feat: Input should capture the Esc button and Clicks while focused [WEB-1251] (#7859)
• 0365ca7 revert: "chore(actors): remove pkg/actors usage from pods.go (#7658) [DET-9652]" (#7908)
• c1a0cf4 chore(actors): remove pkg/actors usage from pods.go (#7658) [DET-9652]
• ff2e16d fix: NTSC use workspace's agent group info (#7892)
• 8bef0d4 chore: no code owners for auto-generated files (#7896)
• 5278758 chore: increase Go's max line length to 120 (#7903)
• b09334d feat: Add display name to user list in cli [MLG-930] (#7901)
• 4301fc2cc feat: move UI related files to the UI kit. (#7852)
• 3f9a980 feat: Hide code related actions based on model definition size (#7854)
• dde10f1 Revert "ci: temporarily move e2e to only nightly [skip ci] (#7837)"
• 35aa028 feat: add an API to get an allocation's exit status (#7731)
• e56ed43 chore: prompt for docs in github question template (#7895)
• 4d827cd fix: remove unused go code (#7893)
• b93f0c9 feat: disable actions of unmanaged experiments/trials (#7874)
• 3b1bebf fix: Metadata deleting last row, cancelling delete [WEB-1655] (#7805)
• 9cddca9 Refactor: Use userSettings store in learning curve (#7783)
• 4a6afb1 feat: add config option to omit default resource pools (#7885)
• e5555ca fix: redefine user columns updated in postgres_users toUpdate (#7890)
• 29561a8 test: quarantine nightly cifar10-keras convergence test (#7780)
• 1cfc7f3 fix: Project move/delete updates UI state [WEB-1668] (#7870)
• 00dfcca test: enable command run tests for hpc (#7880)
• 90d66b8 chore: disable interactive matching for dev bindings (#7747)
• b6632d1 chore: postgres_users.go bun migration [DET-8238] (#7769)
• 826f2b4 feat: containerize performance tests [INFENG-222] (#7863)
• 66f6f4a ci: fix webui test results upload (#7877)
• 64ec5ed Revert "feat: add config option to omit default resource pools (#7696)" (#7878)
• b95d57f fix: use setPartial in experiment list setting (#7873)
• e18d5cf feat: add config option to omit default resource pools (#7696)
• 5e69b6c feat: k8s agent enable disable [DET-9750] (#7779)
• d2e5abb docs: Remove black borders on gif (#7872)
• 144dd0f test: remove deepspeed marks from dsat tests (#7871)
• 40b4341 docs: Add gif to the Readme (#7865)
• a5b29cf docs: Adjust diagrams replacing fluentbit icon (#7867)
• 096935f chore: update user docs (#7864)
• 6d5ad2d docs: Add page for using Determined Agent on Slurm/PBS (#7866)
• 79060ba chore: Remove imagenet (#7664)
• cdceeac feat: Display unmanaged experiments with label (#7861)
• 16a6262 fix: Make models list editing work via ModelActionDropdown [WEB-1603] (#7799)
• 84a6612 fix: doc url in jupyter config modal (#7862)
• 5e04699 fix: fix how we are calling the bert embedding example (#7851)
• d8c7bd2 docs: Clarify meaning of trial api (#7818)
• 813ed36 fix: error message for det agent [enable|disable]. (#7839)
• 9132dcb feat: expose externalExperimentId and externalTrialId (#7840)
• 9874951 chore: bump version: 0.25.1-dev0 -> 0.25.2-dev0
• a5bdfa7 docs: add release notes for 0.25.1 (#7850)
• 38dc440 chore: trial actor refactor (#7821)
• f3aaf4d fix: pass configString once to createexperimentmodal (#7849)
• 346d4aa chore(deps): bump tibdex/github-app-token from 1.8.2 to 2.0.0 (#7847)
• 93e5341 feat: helm ca.cert injection, cluster-wide non-namespaced res creation flag, password change and minor-fix (#7808)
• 442bac6 feat: backend support for inference metric tracking part 2 (#7592)
• 06080b9 feat: allow metrics with duplicate keys and the same value [MLG-890]. (#7820)
• e42c973 feat: enable display of metrics with floating point epoch [MLG-857] (#7829)
• 3c9e0e2 feat: add new API endpoint to get and post accelerator data (#7723)
• febbe18 fix: enable RP bindings management for workspace admins (#7834)
• e843173 ci: temporarily move e2e to only nightly [skip ci] (#7837)
• 4956673 fix: display progress value as it is (#7836)
• 4d74e95 refactor: flipped k8's enable reattach to always true [DET-9726] (#7692)
• 4010b74 chore: nil exception on GetResourcePoolsRequest error (#7835)
• 12d393f fix: dupe checkpoints (#7833)
• 6b12390 fix: log viewer not updating when page switched (#7823)
• 590ea21 ci: fix check-rebaseable syntax [ci skip] (#7826)
• a99385d ci: Add a newline to the output for pre-check (#7824)
• c2ce179 chore: support binary output via dev curl (#7778)
• d78bafe fix: SSO button text color (#7819)
• b4f6f0c fix: correct useResize hook to return proper element sizes [WEB-1656] (#7807)
• e31a077 chore: Split out partial updates into setPartial (#7815)
• 8bcee31 chore: make pre-commit dev setup opt-in. (#7774)
• 675de43 chore: minor copy change (#7810)
• dceb00c chore: agent device discovery too greedy (#7802)
• 1284914 chore(deps): bump actions/checkout from 3 to 4 (#7786)
• 9651c9b fix: progress filter in exp (#7811)
• c284b09 fix: lower severity of allocation log changed when debugging (#7803)
• ff830f9 fix: Learning curve will send falsey metricType (#7809)
• 3c2ab1e chore(deps): bump tibdex/github-app-token from 1.8.0 to 1.8.2 (#7772)
• da23134 docs: HPC launcher doc tweaks, add image scheme docker-archive:// (#7812)
• 1a89c56 docs: Add sections on HPC upgrade and package verificaiton (#7804)
• a51892e fix: Avoid dropdown repeating in ExpList fields dropdown [WEB-1598] (#7800)
• fab413b chore: tools/k8s doesn't use coscheduler (#7795)
• 2b95373 docs: Update the installation guide (#7762)
• 8291a18 ci: quarantine some flaky nightlies (#7725)

0.25.1

Release Notes

0.25.1

Changelog

• 39a421a chore: bump version: 0.25.1-rc2 -> 0.25.1
• 61c11df docs: add release notes for 0.25.1 (#7850)
• e0d0ed2 chore: bump version: 0.25.1-rc1 -> 0.25.1-rc2
• 74eeb77 fix: enable RP bindings management for workspace admins (#7834)
• 1d8e3d2 fix: display progress value as it is (#7836)
• 2c86593 chore: bump version: 0.25.1-rc0 -> 0.25.1-rc1
• 117b173 fix: log viewer not updating when page switched (#7823)
• b93bc72 fix: SSO button text color (#7819)
• cfdacb4 fix: correct useResize hook to return proper element sizes [WEB-1656] (#7807)
• 81b673d fix: progress filter in exp (#7811)
• 29ad1d1 fix: Learning curve will send falsey metricType (#7809)
• b0a7e4e fix: Avoid dropdown repeating in ExpList fields dropdown [WEB-1598] (#7800)
• ebd1906 docs: Update the installation guide (#7762)
• 1bf08e5 chore: bump version: 0.25.1-dev0 -> 0.25.1-rc0
• 7f7e89b chore: lock published urls to preserve redirects
• 59ebdf0 chore: lock api state for backward compatibility check
• 9b6c6c7 fix: Get distributed jobs working with devcluster [FE-181] (#7785)
• d786078 chore: revert trial actor refactor (#7797)
• f4ca02a docs: quick fix for version dropdown (#7796)
• 72d34d9 chore: reduce master log noise (#7794)
• 12a513c chore: Create/document a mechanism to run the nightly tests on a PR [FE-146] (#7750)
• af24954 Revert "chore: track dead code [WEB-258] (#7767)" (#7793)
• c815f76 fix: handles custom TLS certs in enrich_task_logs.py [DET-9803] (#7782)
• 5c83901 chore: remove empty determined/common/api/checkpoint/. (#7776)
• a2b873f chore: suppress the daemonize message on HPC jobs (#7775)
• 5f2f6b8 fix: glitchy width in code editor (#7771)
• bdeb0ea chore: track dead code [WEB-258] (#7767)
• 2167292 chore: trial actor refactor (#7559)
• 06e361e fix: correct date range for avg queued time charts [WEB-1621] (#7754)
• 7c765ae refactor: remove fluent bit & replace with slurm log shipper [DET-9704] (#7639)
• 214198d fix: include unmanaged field in GetExperiment. (#7768)
• f8caa0e feat: Create performance tests [WEB-1458] (#7741)
• b0badb2 fix: Handle chart x-axis with all points at x=0 [WEB-1622] (#7760)
• a6d0fba chore: rearrange log level constants (#7752)
• b22f652 chore: ignore flake8 import restrictions pre-commit check (#7759)
• ef8a295 chore: bump version: 0.25.0-dev0 -> 0.25.1-dev0
• 9333c9d docs: add release notes for 0.25.0 (#7756)
• 8af14ab chore(actors): refactor pod.go (#7617)
• 046e060 test: make error checking case insensitive fixing rbac test (#7749)
• 6c530d3 build: fix go-version-check command (#7751)
• 418931b refactor: make glide-table conform to standard event handler pattern and fix paginated row selection bug [WEB-1471, WEB-1561] (#7704)
• 93d861d chore: use Message for no data in ComparisonView (#7654)
• 4d25428 docs: tweak brew instructions (#7743)
• cf57ce4 chore: upgrade go 1.20 to 1.21 (#7657)
• 0553f19 fix: make rbac messages consistent (#7745)
• e1675b2 fix: not all resource pools should be labeled "default" [WEB-1600] (#7744)
• 2e907ce fix: resource pool card workspace tweaks (#7732)
• 5154c3b fix: proxy tunnel server should use SO_REUSEADDR. (#7735)
• e845836 fix: React build issue (#7742)
• 2fc5f57 fix: Faster polling for first experiment metrics [WEB-1576] (#7740)
• ad46c44 chore: add a new assertion method to check command exit status and report any errors (#7737)
• 418d5ae fix: allow deletion of workspaces when case-insensitive matches exist (#7738)
• 690d451 docs: Reorganize model dev guide sidenav (#7713)
• 9936984 fix: properly display group metrics in metrics tab charts [WEB-1604] (#7727)
• 15e150b fix: allow zeroes for user agent id and group agent id (#7730)
• 8c84750 fix: catch correct import error and set tensorboard logging to false for --test --local (#7715)
• b6f4f30 fix: allow NodeInformer to fail with permission error [DET-9772] (#7703)
• 00a0bc3 fix(cli): not found errors should retain useful context (#7733)
• 688ea88 fix: fix failing e2e_cpu tests (#7734)
• 613e0ce chore: New constructor for Determined objects using existing session. (#7663)
• 108ffea fix: backfilled tasks weren't seen as trial tasks (#7729)
• 1fcd2f9 docs: Add user guides to the Documentation section (#7721)
• d20f577 fix: changing x axis type should reset any current custom zoom (#7728)
• ef5ae83 chore: update determined cli to handle timestamp format for external jobs (#7668)
• 2eadef1 feat: show external jobs on the resource pool page (#7666)
• 2a570bb chore: crash cluster given RM crash (#7621)
• b66ff4a fix: correct GPU name for A100-80GB. (#7724)
• fdddcbf chore: Add nightly tests to release branches (#7720)
• ae6c927 fix: reset chart min/max when changing xaxisdomain (#7719)
• 5e6af2a fix: properly encode metric to keys for LineChart and ParallelCoordinates (#7714)

0.25.0

Release Notes

0.25.0

Changelog

• fea5014 chore: bump version: 0.25.0-rc7 -> 0.25.0
• 3201f27 docs: add release notes for 0.25.0 (#7756)
• 29fbea2 chore: bump version: 0.25.0-rc6 -> 0.25.0-rc7
• 16509f6 test: make error checking case insensitive fixing rbac test (#7749)
• 79a5faa chore: bump version: 0.25.0-rc5 -> 0.25.0-rc6
• 9fde7c7 fix: not all resource pools should be labeled "default" [WEB-1600] (#7744)
• 154168f fix: resource pool card workspace tweaks (#7732)
• d3a42d7 chore: bump version: 0.25.0-rc4 -> 0.25.0-rc5
• 41f9251 fix: React build issue (#7742)
• 1c81e4c chore: bump version: 0.25.0-rc3 -> 0.25.0-rc4
• c4443b3 fix: make rbac messages consistent (#7745)
• 1a83243 fix: allow deletion of workspaces when case-insensitive matches exist (#7738)
• f298ebf fix: properly display group metrics in metrics tab charts [WEB-1604] (#7727)
• 450d1b5 fix: allow zeroes for user agent id and group agent id (#7730)
• fd13b29 chore: bump version: 0.25.0-rc2 -> 0.25.0-rc3
• c121387 chore: bump version: 0.25.0-rc1 -> 0.25.0-rc2
• 1cea783 fix: allow NodeInformer to fail with permission error [DET-9772] (#7703)
• 5d5718f fix(cli): not found errors should retain useful context (#7733)
• 98e0621 fix: backfilled tasks weren't seen as trial tasks (#7729)
• 306baaa fix: changing x axis type should reset any current custom zoom (#7728)
• 406656f chore: bump version: 0.25.0-rc0 -> 0.25.0-rc1
• 00f3af9 fix: correct GPU name for A100-80GB. (#7724)
• 4dd33e4 fix: properly encode metric to keys for LineChart and ParallelCoordinates (#7714)
• a028cd2 chore: Add nightly tests to release branches (#7720)
• 17796e3 fix: reset chart min/max when changing xaxisdomain (#7719)
• 05f808b chore: bump version: 0.25.0-dev0 -> 0.25.0-rc0
• 1cb537e chore: lock published urls to preserve redirects
• 7196181 chore: lock api state for backward compatibility check
• 6e39429 chore: bump version: 0.24.0-dev0 -> 0.25.0-dev0
• 1c8ce3f fix: add missing workspace_id from get_templates (#7706)
• efdc70b fix: code cleanup for mapx unit test (#7710)
• 0a34529 fix: add unit test cases for mapx methods Values and Clear (#7699)
• 1a4bee4 feat: det deploy gcp support for a2-ultragpu and g2-standard. (#7702)
• 2fd2535 fix: users can see inaccessible RPs (#7707)
• e83660c fix: rp bindings intg test failure (#7701)
• f1e9b72 chore: Remove estimatortrial (#7700)
• e2f2173 feat: replace clone function with structuredClone and add polyfill (#7624)
• f528cc6 fix: botched rebase/rename in the detached mode. (#7695)
• 08de858 fix: Continue Trial modal does not reset mode [WEB-1566] (#7688)
• f9c5600 fix: error message in jupyter (#7693)
• 2607fd1 fix: patch workspace has duplicate update statements (#7697)
• bf1b87b fix: correct outstanding error in mapx (#7698)
• cdc41e9 fix: add type check to pod spec merge (#7691)
• e340d50 chore: add Values and Clear methods for mapx (#7669)
• 6bc3c68 docs: algolia scraper to scrape only xml (#7690)
• 0977986 docs: fix new release notes (#7694)
• 76134b2 chore: dev cli support for calling master apis (#7462)
• 43c715b docs: add release notes for 0.24.0 (#7680)
• c513e70 chore: add new RBAC permission view external jobs (#7671)
• 128b106 docs: work around bug causing version dropdown to fail (#7685)
• 26559ed feat: check if default resource pools are bound (#7687)
• 7d8fce5 docs: improve writing of the github readme (#7689)
• 857309f feat: add rp bindings permissions (#7673)
• 5c68568 chore: api intg tests [DET-9725] (#7589)
• 1ad812d docs: Improve the GitHub Readme (#7613)
• 8239b19 fix: default pools editable and submittable (#7647) (#7672)
• 4f60d64 chore: unpin click version (#7684)
• ec90842 chore(deps): bump arduino/setup-protoc from 1 to 2 (#7537)
• b4cbe9c chore: enable mask closable by default for drawers (#7676)
• b1f02ed chore: limit reported slots (#7683)
• 75c1f17 fix: tensorflow version for macos (#7679)
• 0a5b406 fix: allow special characters in user manangement filter (#7681)
• 1af32e3 feat: Update user.modified_at when user added or removed from groups (#7665)
• 23a8224 fix: Case-insensitive client-side username search [DET-9770] (#7677)
• a368a55 chore: limit reported slots (#7648)
• 60a07e4 fix: make -C master clean build [DET-9333] (#7660)
• fcf7807 chore: custom metrics group in new experiment list (#7518)
• dc006b1 docs: Fix formatting (#7670)
• 6ee0521 docs: Introduce users to pachyderm w det (#7661)
• 4ef8b20 feat: detached mode v1 / core api v2. (#7060)
• d580ecf fix: allow checkpoints to be GCed without validation metrics and add tests (#7653)
• 204caa5 fix: optional chaining in extractMetricValue (#7662)
• 0ead6ac chore: telemetry actor refactor [DET-9663] (#7585)
• e437bd8 docs: Point to pytorch distributed launcher (#7649)
• 08ff4be docs: fix epoch metrics article (#7643)
• 395aa40 docs: Update resource pool to workspace mapping (#7642)
• 28cbdc2 fix: properly show the pagination for experiment list paged view (#7638)
• 68624f6 chore: avoid creating new table columns for non-legacy metrics (#7656)
• e558063 chore: Add eslint rule for imports to take one line [WEB-1567] (#7650)
• 0bc5dce ci: bump everything to torch==1.11 (#7599)
• 4eb6ebe fix: Project delete/move triggers update of workspace projects list [WEB-1497] [WEB-1377] (#7646)
• 145dd63 ci: indicate GHA run URL when reporting a cherry-pick conflict (#7635)
• 8d2b531 chore: Show Tooltip instead of actions for Default Resource Pools [WEB-1554] (#7644)
• 0bab558 fix: select component width (#7640)
• 32fac33 chore: use eslint rule to avoid relative imports through parent [WEB-1496] (#7637)
• dfd5475 chore: remove unused parseFloat for decoding string metric values (#7641)
• e472ea7 fix: alphabetical binding workspaces and search copy change [WEB-1552, WEB-1553] (#7633)
• 7b96933 fix: properly clear out the settings from the database [WEB-1559] (#7636)
• e9e66b1 fix: fix incorrect return type for downsampled metrics (#7618)
• 930fc9d feat: custom metric groups (formally known as types) [WEB-1469] (#7570)
• 75e93d9 docs: bump rstfmt version (#7611)
• 34c5b5a fix: trigger jobs fetchAll on pagination changes [WEB-1546] (#7602)
• 29e63af Remove say workaround and update version (#7628)
• 0a24176 chore: fix pod-spec merge logic (#7574)
• ce3136a feat: Don't show charts where all series are Loaded(no data) [WEB-1524] (#7609)
• d3c027b feat: OptionsMenu moved to left group (#7623)
• 4d002c4 docs: Add article on how to view epoch metrics (#7504)
• 4327a25 fix: rp binding resolving resource pools (#7629)
• f79e95e ci: fix release branch selection when cherry-picking EE PRs (#7630)
• d6a5c79 fix: Handle metric names finish loading, but still empty (#7634)
• 2844566 chore: support mobile view in UIKit [WEB-1314] (#7626)
• 3ec9c49 fix: button filter text (#7632)
• 0d293b5 fix: ChartGroup vertical spacing (#7631)
• 6fd4b21 feat: replace custom isEqual to lodash isEqual (#7625)
• fa91629 feat: add searcher metric sorting (#7614)
• 5ff0b7d fix: avoid converting workspace name to sentence casing [WEB-1548] (#7622)
• 37259d4 fix: treat searcher metrics value as a number in the ui (#7612)
• bbe70e0 feat: Resource pool tab for workspace (#7582)
• 369ddf3 feat: Copy cell value from experiment list table (#7604)
• 3cb434d fix(actors): trial lifetime must contain allocation lifetime, still (#7615)
• 033a9f6 fix: Single-point tooltip closes when mouse exits chart [WEB-1541] (#7595)
• b8f95ad docs: Add css rule to turn off scrolling when clicking on section links (#7610)
• 9994aa3 fix: code editor height issues (#7573)
• acdd6c4 docs: improve a release note (#7601)
• 37cc9f0 chore: add agent --image-root (#7597)
• 69ab985 refactor: trial's can have one or many tasks [DET-9647] (#7355)
• 7c076b2 ci: fix remote name in PR tracking script (#7607)
• b2ee9b3 fix(actors): create valid fake group actor for checkpoint GC, don't leak it (#7606)
• 776be10 fix: fix checkpoint gc which was incorrectly deleting some checkpoints (#7523)

0.24.0

Release Notes

0.24.0

Changelog

• 620162e chore: bump version: 0.24.0-rc5 -> 0.24.0
• 809eda9 docs: add release notes for 0.24.0 (#7680)
• 399ad9c chore: bump version: 0.24.0-rc4 -> 0.24.0-rc5
• b1b2bf2 fix: allow checkpoints to be GCed without validation metrics and add tests (#7653)
• 378fe8d chore: bump version: 0.24.0-rc3 -> 0.24.0-rc4
• c1ffb75 docs: fix epoch metrics article (#7643)
• abf253b make error message more accurate (#7659)
• d57c5ec fix: default pools editable and submittable (#7647)
• b7aa1f4 chore: bump version: 0.24.0-rc2 -> 0.24.0-rc3
• 9095ee8 fix: properly clear out the settings from the database [WEB-1559] (#7636)
• 328ff9a docs: Add article on how to view epoch metrics (#7504)
• 39fa1db fix lint from #7629
• 24b89f7 fix lint from #7634
• 8806cd5 fix: Handle metric names finish loading, but still empty (#7634)
• a104da8 Revert "fix: Handle metric names finish loading, but still empty (#7634)"
• ed5e38c fix: rp binding resolving resource pools (#7629)
• aafcff8 fix: Handle metric names finish loading, but still empty (#7634)
• 6846471 fix: treat searcher metrics value as a number in the ui (#7612)
• 7aea1b4 fix(actors): trial lifetime must contain allocation lifetime, still (#7615)
• 524cc58 chore: bump version: 0.24.0-rc1 -> 0.24.0-rc2
• 1f0b33e fix: fix checkpoint gc which was incorrectly deleting some checkpoints (#7523)
• 57d7bc4 fix(actors): create valid fake group actor for checkpoint GC, don't leak it (#7606)
• 3a20df1 chore: bump version: 0.24.0-rc0 -> 0.24.0-rc1
• d17dbfd chore: bump version: 0.24.0-dev0 -> 0.24.0-rc0
• d7b2f5c chore: lock published urls to preserve redirects
• 4931751 chore: bump version: 0.23.5-dev0 -> 0.24.0-dev0
• bfd964c fix: proto user always has false remote (#7126)
• 5bc1666 chore: fix incorrect make invocation (#7605)
• e960b57 Update README.md (#7587)
• c957354 chore: bumpenvs 24.0 (#7594)
• 3c44a2f fix: ensure selected file path matches loaded file in codeeditor (#7563)
• 84a5800 fix: fetch the latest projects (#7598)
• 50fc3e7 chore: update comment to clarify endpoint behavior (#7583)
• 38a49cf docs: update doc string related to adding enable_tensorboard_logging flag (#7600)
• 549273c feat: Enable disabling Tensorboard logging [MLG-22] (#7508)
• 6d6d5ad chore: Remove ptl adapter (#7591)
• 306e0df fix: remove flaky component if new xp list is active. (#7590)
• db7ae17 ci: Add pytorch2 tests (#7581)
• 62cc08c feat: Move new charts into always-on Metrics tab [WEB-1522] [WEB-1523] (#7542)
• 5cf569d docs: fix lint for docs/architecture/introduction.rst (#7586)
• 7c6aead chore: migrate to singularity --nvccli (#7576)
• 923a742 chore: Make explist_v2 generally available (#7561)
• efc4458 Added Profiling to the Benefits table in the intro (#7580)
• 4ad4a76 fix: quick disambiguation on exp. checkpoint size (#7579)
• d40ae77 chore: fix rph docs url publishing step (#7487)
• 5f21741 docs: edit release notes readme (#7578)
• cd19761 fix: Button icon spacing (#7568)
• 22ca050 feat: get unbound pools endpoint [DET-9696] (#7527)
• d05e530 test: unpinning responses (#7551)
• d63d903 chore: Parse and format APIExceptions (#7531)
• ada554f docs: edit some release notes (#7540)
• 9d44ffc docs: add sphinx-tabs extension (#7577)
• 3623a23 docs: Edit the release notes readme (#7572)
• 04fb7ef chore: summarize state of mock_client_test.go [DET-9731] (#7571)
• 708db2d fix: Accomodate partial experiment list settings (#7564)
• 345da9c chore: add linter for google style guide-compliant python imports. (#7550)
• 2969f53 chore(rm): refactor cluster management APIs into RM (#7569)
• f10b474 chore: update type to handle generic summary metric types [WEB-1538] (#7567)
• 577cebe chore: Add web as codeowner for /webui (#7565)
• 56a59c0 fix: hide cluster logs in mobile view (#7557)
• 4bcb04b fix: log level filter broken on det t/e logs [MLG-798] (#7558)
• 88338c1 fix: overwrite bindings not working for zero length list (#7549)
• 3258b92 fix: correct name for external jobs (#7566)
• fd7edc8 fix: scrolling when dragging column headers in Glide Table (#7548)
• fbcc4b5 fix: Dont seek min and max on projects with 0 experiments (#7560)
• 6f448b4 feat: Replace sum and count training metrics with mean in new experiment list (#7493)
• de9e4f5 chore(actors): refactor checkpoint GC tasks (#7435)
• da4696b chore: master linting less verbose (#7553)
• 99279ad feat: Provide an interface to enable resource managers to show External jobs on the resource pool queue. (#7070)
• 42feb36 docs: apply minor tweaks (#7554)
• 3534b49 docs: correct some det deploy gcp docs facts. (#7556)
• f42cdcc chore: add copy to manage bindings modal (#7555)
• 88a93b0 chore: update default exp list columns [WEB-1488] (#7534)
• cc8cb63 fix: Groups modal does not include inactive users [WEB-1256] (#7528)
• b693ddd build: convert svgs to react by default (#7541)
• 01bdf93 feat: Heatmap support for glide table (#7267)
• 277e124 chore: update fmt-sql config and version (#7544)
• db8e965 ci: setup CODEOWNERS for ml-sys team. (#7546)
• 318ffda chore: Add new update function to userSettings store (#7469)
• bde72d6 chore: use Dropdown for Experiment List menus (#7522)
• fcbc80e refactor: add Spinner to UI Kit [WEB-1451] (#7498)
• 03f503c ci: add a sql fromatter (#7538)
• ec32f2f ci: handle duplicate cherry-picks of a PR to release branch (#7502)
• 3db61f5 fix: compare charts showing no data while loading [WEB-1485] (#7516)
• 43ff696 fix: Improve HPC error shutdown to improve logging [FE-44] (#7488)
• a69ec27 chore: temporarily rollback agent usage of --nvccli (#7533)
• 90e131a fix: fix default pools and refactor (#7535)
• e86d47a fix: use proper experiment project resolution (#7532)
• c7ac817 fix: disable Manage Bindings option for default pools [WEB-1521] (#7519)
• 5f0ff57 fix: custom proxies do not work for trials in slurm [DET-9718] (#7529)
• 4722ee9 fix: filter scrollbar adjustment (#7530)
• 9867d4d fix: Selected experiments in glide table persist [WEB-1366] (#7289)
• 39013e9 chore: add option to syntax highlight cli json output (#7471)
• e75c7f4 chore: rename db metric references to custom_type (#7473)
• fa626b0 chore(actors): refactor allocation actor without actors (#7391)
• 303f0d2 chore: bump version: 0.23.4-dev0 -> 0.23.5-dev0
• 630f721 docs: add release notes for 0.23.4 (#7524)
• 2dd26d5 fix: Don't return a workload for deleted checkpoints [WEB-1505] (#7491)
• e13190a fix: Reset cluster jobs pagination when offset is out of bounds (#7521)
• ab21b6e chore: rollback torch 1.7 support removal. (#7525)
• 56d444e chore: update vite (#7505)
• ec0e750 style: update copy and add dividers to exp list table action dropdown [WEB-1490] (#7506)
• cddaf4f fix: rename checkpoints (#7513)
• 7b6f8e1 chore: remove double newline in cli error messages (#7472)
• 44f03fd fix: det tunnel should work with proxy port exposed [FE-121] (#7492)
• 28d18d9 fix: unbumpenvs. (#7496)
• d868241 feat: Pytorch2 necessary changes (#7515)
• 54c5a7d style: update experiment selection label [WEB-1510] (#7510)
• 0c4a19f fix: rp-workspace mapping RP not found [WEB-1508] (#7514)
• 9deeb48 fix: support searcherMetric (#7511)
• 525ebfe chore: consolidate k8s informers code & fix Makefile mocks (#7455)
• 5ef7d30 ci: put all GHA jobs for release tracking in a concurrency group (#7507)
• 47ac573 fix: Metrics with dot in name appear correctly in trial view [DET-9691] (#7450)
• 856a963 fix: k8s determined-container gets wrong RunAsUser (#7503)
• 18821cd ci: avoid latest responses==0.23.2 (#7501)
• c02fce2 docs: rp workspace mapping release notes (#7499)
• 33ab05a fix: typo that allows binding default aux pool (#7500)
• 4fde629 fix: avoid crashing the new exp page (#7489)
• 28953dc docs: FE-120: Add job_history_enable = True to PBS installation requirements (#7480)
• fcfcbe6 docs: Add RP to Workspaces user guide (#7326)
• 707221c docs: Describe WebUI settings (#7478)
• 8366613 chore: add migration number validation to migration util (#7347)
• 57c3e8c test: skip test_efficientdet_coco_pytorch_const. (#7494)
• 8471017 test: fix rbac test failures (#7476)
• aac4272 fix: stop showing invalid loading (#7470)
• 9327830 fix: ChartGrid styling (#7485)
• 7f35b7d fix: rp workspace mapping not working (#7490)
• d29d49d feat: backend support for inference metric tracking part 1 (#7375)
• db01c03 fix: chart tooltip overflow (#7484)
• 99ac604 fix: experiment list compare panel resize (#7477)
• b11d1cb fix: only let cluster admins manage resource pool bindings [WEB-1476] (#7483)
• 6f147fd docs: Add torch batch process example (#7482)
• b71e79a fix: rename checkoutCount to checkpoints (#7481)
• ebac1d3 ci: Conda bump (#7479)
• 1e021bc ci(aws): fix RDS connections (#7475)
• 19945b6 ci: correctly change item status in release tracking (#7460)

0.23.4

Release Notes

0.23.4

Changelog

• f5484dd chore: bump version: 0.23.4-rc4 -> 0.23.4
• 9c8ea0c docs: add release notes for 0.23.4 (#7524)
• 91e6b82 chore: bump version: 0.23.4-rc3 -> 0.23.4-rc4
• 8afa912 fix: rename checkpoints (#7513)
• d8d671b chore: bump version: 0.23.4-rc2 -> 0.23.4-rc3
• 6b26541 fix: unbumpenvs. (#7496)
• fdc9acf fix: rp-workspace mapping RP not found [WEB-1508] (#7514)
• 99f9cc3 fix: support searcherMetric (#7511)
• 65f76b2 chore: bump version: 0.23.4-rc1 -> 0.23.4-rc2
• dc4472e fix: k8s determined-container gets wrong RunAsUser (#7503)
• 54b6602 ci: avoid latest responses==0.23.2 (#7501)
• ecbfe61 docs: rp workspace mapping release notes (#7499)
• 3131f43 fix: typo that allows binding default aux pool (#7500)
• 593ace0 fix: avoid crashing the new exp page (#7489)
• 557a272 docs: Add RP to Workspaces user guide (#7326)
• 7fcff5a test: skip test_efficientdet_coco_pytorch_const. (#7494)
• 8648e05 test: fix rbac test failures (#7476)
• 17ac0d6 chore: bump version: 0.23.4-rc0 -> 0.23.4-rc1
• 8db5076 fix: rp workspace mapping not working (#7490)
• ecaad56 fix: rename checkoutCount to checkpoints (#7481)
• ed19b69 ci: Conda bump (#7479)
• 66f67f9 ci(aws): fix RDS connections (#7475)
• 2d64e16 chore: bump version: 0.23.4-dev0 -> 0.23.4-rc0
• 4dda861 feat: RP<>workspace mapping (#7461)
• 9bffd21 chore: add a toy experiment example for pushing generic metrics (#7442)
• 7e7ac8b fix: use friendly names for user settings (#7465)
• fda7235 chore: Turn new experiment listing off by default (#7466)
• e6cabd6 docs: rephrase the home page title (#7463)
• 4e9748f feat: Allow user setting of feature flags (#7438)
• 307e2ee chore: avoid postgres uuid extension (#7459)
• a66db23 chore: explist_v2 feature switch should default to on (#7249)
• 3ba9904 style: mobile version of exp list v2 [WEB-1422] (#7433)
• 6fe9218 test: mark RBAC-related tests 'e2e_cpu_rbac' (#7401)
• aafac46 fix: chart axis long label (#7451)
• 9e37ed4 docs: Reset home page tiles and sidebar (#7446)
• 8dc0cfd fix: wider width for top trials select (#7431)
• b67d981 ci: fix latest release branch selection (#7432)
• c915f84 chore: podspec-capability-bugfix (#7447)
• 2cc2656 feat: Pin column when compare panel open (#7414)
• 04485b2 chore: update docs hpe compliance web section id (#7449)
• 78d8baa feat: minor edit to torch batch process embedding example (#7440)
• 2ff3d2e chore: allow no pinned columns [WEB-1423] (#7419)
• 210c446 feat: sort resource pools by name when creating new jupyterlab sessions [WEB-1138] (#7444)
• 2e776a4 fix: chart tooltip has exp. name (#7410)
• bd470b6 feat: Merge data and row height menus in explist v2 (#7405)
• 9caaae8 docs: update aria labelling for sidebar toggles and toctree groups (#7381)
• 9de58bd fix: update get agents call for agent disable all (#7441)
• 2d3120d feat: master audit logging should log failed requests at Info level. (#7434)
• 34347a1 feat: Compare icon reflects status (#7443)
• c409c31 style: apply hover color for selected table rows (#7429)
• 639e34a fix(cli): compound keys in --config opts for commands (#7439)
• 70e65df feat: allow-pachyderm-notebook-extension [DET-9355] (#7395)
• bfdccdf docs: Add documentation about passing in optional tensorboard arguments [MLG-333] (#7352)
• 07e12f7 fix: table column copy change (#7373)
• 8300458 chore: events & preemption listener actor refactor [DET-9617] (#7256)
• 497df17 feat: [MLG-647] Add batch inference examples using Core API and torch_batch_process (#7274)
• 119f577 chore: fail with non-zero exit code when password change fails (#7403)
• 06c7d0b feat: edit and reset raw user settings [WEB-1361] (#7377)
• f98d15a fix: perf improvement in useTrialMetrics (#7426)
• 2ce5aaf fix: clear table selection when changing pages (#7418)
• 533f3a7 chore: migrate to singularity --nvccli [DET-9081] (#7337)
• 1002645 ci: fix e2e tests for tf keras cifar10 example (#7422)
• 71243e3 chore: agent provisioner refactor (#7287)
• d82f35f fix: Correctly sequence updates in useSettings bridge code (#7423)
• 0fc6238 feat: Add Bert embedding torch_batch_process example (#7402)
• 22f82a1 feat: obfuscates slot id in agent summary(#7421)
• 0855b78 feat: Settings drawer (#7356)
• 6da387b fix: fix: label default compute and default aux pools differently (FE-108) (#7420)
• e2112c9 chore: k8s pod & node informer refactor (#7182)
• c20ad5a fix: show loading state when applicable for comparison view tabs (#7370)
• 30a20ba fix: All metrics viewable in the glide table compare tab [WEB-1372] [WEB-1394] (#7412)
• 8ceb955 docs: remove RBAC limitations section (#7415)
• 403754b ci: warn about PRs that are from forks but by users with write access (#7380)
• ee68c3d perf: migrate checkpoint v1 into v2 table (#7325)
• 3e9eaed chore: skip process auth admin check if rbac is enable (#7400)
• af9d562 ci: make CASPER_TOKEN optional in track-pr script (#7379)
• c510035 chore: bump version: 0.23.3-dev0 -> 0.23.4-dev0
• 37e6d9c docs: add release notes for 0.23.3 (#7413)
• b795a23 fix: custom column resize [WEB-1341] (#7365)
• 227a684 fix: replace div with table in Trial Comparison View (#7341)
• 59f34bd chore: update data loading in Keras CIFAR10 example (#7378)
• 9464a1f chore: change compare tab refresh behavior [WEB-1434] (#7397)
• f7a6ba4 chore: add release notes for notebook tls (#7408)
• 6ef9862 docs: hpe docs compliance for section 5 [WEB-1292] (#7383)
• 3718a4f chore: update left over api references to metric type (#7399)
• 4fdf5ae fix: userSettings store should function correctly when useSettings is active (#7409)
• 5ae6311 chore: Workspace as an object in the SDK. (#7387)
• 638f56e chore: add generic metrics support to harness (#7407)
• 60987a5 fix: refactor checkpoint modal so it closes correctly [WEB-1441] (#7398)
• fccd378 ci: Fix GKE cluster version used for CircleCI (#7385)
• 28f2c03 fix: add fp16 flags to hf ds examples (#7265)
• 2ac22f0 fix: install sigusr1 on main thread only. (#7350)
• 03c3faf fix: memoize settings in model detail and experiment detail pages (#7390)
• 056d414 chore: Clean up redundant aria-labels, correct wrong aria-labels [WEB-1379] (#7384)
• 1f5b9a0 fix: rbac filter on columns API (#7386)
• a0a87cd feat: add update master config API end point with RBACK and CLI command (#7318)
• 839dcbd chore: bump up flake8 to 3.9.2 (#7389)
• 7eaea60 chore: rename metric type concept to metric group (#7353)
• dd15be9 fix: add condition by default in filter (#7362)
• fe85366 fix: new exp detail crash [WEB-1444] (#7382)
• 4e80c72 fix: Metrics reporting no data (#7368)
• b9d820e fix: make checkpoint_count explicit (#7374)
• 2d66b53 ci: fix some bugs in tracking of cherry-picked PRs (#7367)
• 13a6ad8 chore: clean up comments in trial and metric code (#7376)
• 5aed4c5 chore: update make_url utility with fallback (#7259)
• 03bb110 chore: use det custom json encoder in print_json (#7371)
• 4f92cab fix: use DEFAULT_COLUMNS in GroupManagement table (#7366)
• bea92ef feat: add training metrics to columns api (#7320)
• b2fadf1 fix: Trials use their own trial state enum [DET-9639] (#7354)
• 90257d6 feat: allow merging metrics with same batch number (#7304)
• 5f18070 ci(circle/test-unit): regroup gpu unit tests (#7345)
• c2a8113 fix: handle long values (#7349)
• 4c41432 docs: add HPE marketing analytics code (#7269)
• 223f28d fix: TypeError 'NoneType' object processing exp config (#7290)
• 43abf42 fix: Fix issue in user settings store (#7363)
• 1640d7a fix: reverting not-found-errs change in python user-groups (#7361)
• 6e0c2bf chore: bumpenvs (#7348)