Releases: GrayboxTech/weightslab
Releases · GrayboxTech/weightslab
Release list
Release v1.4.0
╭─────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │
│ $$\ $$\ $$\ $$\ $$\ $$\ $$\ │
│ $$ | $\ $$ | \__| $$ | $$ | $$ | $$ | │
│ $$ |$$$\ $$ | $$$$$$\ $$\ $$$$$$\ $$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ │
│ $$ $$ $$\$$ |$$ __$$\ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____|$$ | \____$$\ $$ __$$\ │
│ $$$$ _$$$$ |$$$$$$$$ |$$ |$$ / $$ |$$ | $$ | $$ | \$$$$$$\ $$ | $$$$$$$ |$$ | $$ | │
│ $$$ / \$$$ |$$ ____|$$ |$$ | $$ |$$ | $$ | $$ |$$\ \____$$\ $$ | $$ __$$ |$$ | $$ | │
│ $$ / \$$ |\$$$$$$$\ $$ |\$$$$$$$ |$$ | $$ | \$$$$ |$$$$$$$ |$$$$$$$$\ \$$$$$$$ |$$$$$$$ | │
│ \__/ \__| \_______|\__| \____$$ |\__| \__| \____/ \_______/ \________| \_______|\_______/ │
│ $$\ $$ | │
│ \$$$$$$ | │
│ \______/ │
│ │
│ Inspect - Edit - Evolve Neural Networks │
│ │
╰─── By GrayBx, v1.4.0 ───────────────────────────────────────────────────────────────────────────────╯
Weightslab
Follow us on LinkedIn | Graybx
Welcome
Welcome to the v1.4.0 release of WeightsLab. This release was made with the idea to improve how you Inspect Data, Edit and Evolve Neural Networks, all by Graybx:
- PR #266: > Re-opened from #265 with the Ultralytics example notebooks excluded (kept on
devonly, not ready for this release). ## WeightsLab v1.4.0 — Release Notes ### 🚀 Highlights - Collab compatibility improvements - PyTorch notebooks support - UL notebooks integrations - Dependency fixes (notably aroundnumpy > 2andprotobuf) - Docker dependency cleanup - Removed Docker-related dependencies/repositories - Updated UI Docker bridge (including rename) - Updated README and docs accordingly - **Model...
One platform. Unmatched flexibility. Ultimate control.
Make sure to install our last version on PyPI
Happy Training!
What's Changed precisely:
#266 v1.4.0
5eda2fcadd new nworkers tests for perfsa4cf489Stop DataLoaderInterface from blocking on pause_controller during ite…580c80fMove pause_at_step trigger from DataLoaderInterface into GuardContext41b76b8Remove dead DataLoaderInterface._wait_if_pausedd0577d0ws-detection: default num_workers=2, add num_workers benchmark harness8c501f8Drop now-unused threading/pause_controller imports from dataloader_in…9f23295ws-detection: drop bench.py/run_bench.sh from the branch5ec83e5Revert ws-detection example changes; keep this branch framework-onlyd240dffFix ws-detection example: restore runnable training loop69edb2bfix oom bug on break by slice (#186)b799aa8179 audit and export user interactions success and failure to json an…b210dd9184 dataframe granularity (#187)8f8d5f1Ddp support wip (#185)416774fFix lint code quality issue58f72b3Fix ruff and vulture code quality issue834ac7ffix utests after mergingf6d7904Code quality issuese43aa1bFinal code quality checkeef7a21change ci code quality as important for CI1557f9fFix dev release CI9953476update release note formatbed3e0fImprove release notes: custom template with PR links, contributor pro…e00bd5cFinal release note fix1258ebdfix instance_ids bugs matching for batch target format (targets=[bb_A…4c88585188 logger granularity and structure (#189)7cf5c0crevert merging PR #185861911cFix bug from reverted Merged Conflicts24bf58fremove hardcoded path from config3383035fix iou metric name685c88ffix merged conflicts error on fct name10a65ffRemove hardcoded value from src ModelInterface456ffb1fix merged conflicts function namef411225Add weightslab.integrations.ultralytics SDK (#191)6cc624bfix useless imports and vars: code quality issuesc5f0807Force val to letterbox to (imgsz, imgsz) like train1974d3bSource val rect from ledger config; default False901da04small fixes around certs. and windows; and add ul deps in wlca8e38bfix ul deps versionf3d5128fix quotes issue3459058Add geometry test for val rect-mode opt-in12c91efRevert "Add geometry test for val rect-mode opt-in"ed2d64bFix failing unit tests and dataloader infinite loopa9a2165Rework UI onboarding CLI: secure-by-default launch, cleanup, banner helpaae4b65Make CI run the real test suite and fix/triage dev test failuresbea6314Make WEIGHTSLAB_CERTS_DIR the single source of truth for TLS/authe985a66Keep a valid certs bind-mount in unsecured/--no-certs launches6a09909Addweightslab example startto run the bundled classification example314c60cfix ui docker bridge and simplify certsf40792fadd torchmetrics deps for cls example2e4778dfix onboarding certs issue and update doc and readme912c03cFix certs issues and related utestsaf6846cFix certs issues; change cls config; and fix issues with hist and age…579d8ccRefactor wl examples syntax0c7f1c0fix permissions issues and docker v18695667fix exec bits from exec files12f5130fix cmds tests57c8313Integrate 2d 3d point cloud detection usecase with Lidar (#199)5432688fix code quality6ed796ffix code quality0927cceAdd missing source files skipped by data/ gitignore rule1e23d37Fix code quality issue8076c25fix utests73fca0cFix 4 ui_docker_bridge unit tests broken by --certs flag refactor19f2f16Exclude heavy blob columns from histogram/metadata sweeps1bbd971Python version tests in workflows67f90a5remove useless certs8deb810update gitignore and add debug print exce38c4f4add docker initialization materials72d682efix logger prints format and spam3b88c59maximum resolution set to 360p by default6c3e1d5add auto start in examples59a45bcFix main process issues with windows and logs; and fix ultralytics in…117bb4eremove useless dirff662f2Disable the WatchDog; fix evaluate feature in nograd mode and no guar…0bee476Fix code quality issue7b06e0fFix UL issue with custom EMA and evaluation feature58ce5eeAllow to run CI from custom branch with correct trigger flag in commit4e84dc0Fix pandas incompatible-dtype FutureWarning in dataframe merge/upsert997e914Add agent memory to repo for user7698e20fix prediction normalization issuesee9d6e6Fix ledgered parameters issues61bc12aFix new exp. hash after pause 4 the first timeeca9e25[skip ci] fix parameters wrapping in examples3751825V1.3 UI optimization and new features (#214)
Thank you!
Dev Release v1.4.0.dev3
╭─────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │
│ $$\ $$\ $$\ $$\ $$\ $$\ $$\ │
│ $$ | $\ $$ | \__| $$ | $$ | $$ | $$ | │
│ $$ |$$$\ $$ | $$$$$$\ $$\ $$$$$$\ $$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ │
│ $$ $$ $$\$$ |$$ __$$\ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____|$$ | \____$$\ $$ __$$\ │
│ $$$$ _$$$$ |$$$$$$$$ |$$ |$$ / $$ |$$ | $$ | $$ | \$$$$$$\ $$ | $$$$$$$ |$$ | $$ | │
│ $$$ / \$$$ |$$ ____|$$ |$$ | $$ |$$ | $$ | $$ |$$\ \____$$\ $$ | $$ __$$ |$$ | $$ | │
│ $$ / \$$ |\$$$$$$$\ $$ |\$$$$$$$ |$$ | $$ | \$$$$ |$$$$$$$ |$$$$$$$$\ \$$$$$$$ |$$$$$$$ | │
│ \__/ \__| \_______|\__| \____$$ |\__| \__| \____/ \_______/ \________| \_______|\_______/ │
│ $$\ $$ | │
│ \$$$$$$ | │
│ \______/ │
│ │
│ Inspect - Edit - Evolve Neural Networks │
│ │
╰─── By GrayBx, v1.4.0.dev3 ──────────────────────────────────────────────────────────────────────────╯
Weightslab
Follow us on LinkedIn | Graybx
Welcome
Welcome to the v1.4.0.dev3 release of WeightsLab. This release was made with the idea to improve how you Inspect Data, Edit and Evolve Neural Networks, all by Graybx:
(no pull requests found since last release)
One platform. Unmatched flexibility. Ultimate control.
Make sure to install our last version on TestPyPI
Happy Training!
What's Changed precisely:
dd56007release: drop the ls-remote probe for weights_studio's dev branch
Thank you!
- GuillaumePELLUET
Dev Release v1.4.0.dev2
╭─────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │
│ $$\ $$\ $$\ $$\ $$\ $$\ $$\ │
│ $$ | $\ $$ | \__| $$ | $$ | $$ | $$ | │
│ $$ |$$$\ $$ | $$$$$$\ $$\ $$$$$$\ $$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ │
│ $$ $$ $$\$$ |$$ __$$\ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____|$$ | \____$$\ $$ __$$\ │
│ $$$$ _$$$$ |$$$$$$$$ |$$ |$$ / $$ |$$ | $$ | $$ | \$$$$$$\ $$ | $$$$$$$ |$$ | $$ | │
│ $$$ / \$$$ |$$ ____|$$ |$$ | $$ |$$ | $$ | $$ |$$\ \____$$\ $$ | $$ __$$ |$$ | $$ | │
│ $$ / \$$ |\$$$$$$$\ $$ |\$$$$$$$ |$$ | $$ | \$$$$ |$$$$$$$ |$$$$$$$$\ \$$$$$$$ |$$$$$$$ | │
│ \__/ \__| \_______|\__| \____$$ |\__| \__| \____/ \_______/ \________| \_______|\_______/ │
│ $$\ $$ | │
│ \$$$$$$ | │
│ \______/ │
│ │
│ Inspect - Edit - Evolve Neural Networks │
│ │
╰─── By GrayBx, v1.4.0.dev2 ──────────────────────────────────────────────────────────────────────────╯
Weightslab
Follow us on LinkedIn | Graybx
Welcome
Welcome to the v1.4.0.dev2 release of WeightsLab. This release was made with the idea to improve how you Inspect Data, Edit and Evolve Neural Networks, all by Graybx:
(no pull requests found since last release)
One platform. Unmatched flexibility. Ultimate control.
Make sure to install our last version on TestPyPI
Happy Training!
What's Changed precisely:
ffdc05eFixing WL Release CI5186d34release: surface why the weights_studio dev-branch probe falls back to main
Thank you!
- GuillaumePELLUET
Dev Release v1.4.0.dev1
╭─────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │
│ $$\ $$\ $$\ $$\ $$\ $$\ $$\ │
│ $$ | $\ $$ | \__| $$ | $$ | $$ | $$ | │
│ $$ |$$$\ $$ | $$$$$$\ $$\ $$$$$$\ $$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ │
│ $$ $$ $$\$$ |$$ __$$\ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____|$$ | \____$$\ $$ __$$\ │
│ $$$$ _$$$$ |$$$$$$$$ |$$ |$$ / $$ |$$ | $$ | $$ | \$$$$$$\ $$ | $$$$$$$ |$$ | $$ | │
│ $$$ / \$$$ |$$ ____|$$ |$$ | $$ |$$ | $$ | $$ |$$\ \____$$\ $$ | $$ __$$ |$$ | $$ | │
│ $$ / \$$ |\$$$$$$$\ $$ |\$$$$$$$ |$$ | $$ | \$$$$ |$$$$$$$ |$$$$$$$$\ \$$$$$$$ |$$$$$$$ | │
│ \__/ \__| \_______|\__| \____$$ |\__| \__| \____/ \_______/ \________| \_______|\_______/ │
│ $$\ $$ | │
│ \$$$$$$ | │
│ \______/ │
│ │
│ Inspect - Edit - Evolve Neural Networks │
│ │
╰─── By GrayBx, v1.4.0.dev1 ──────────────────────────────────────────────────────────────────────────╯
Weightslab
Follow us on LinkedIn | Graybx
Welcome
Welcome to the v1.4.0.dev1 release of WeightsLab. This release was made with the idea to improve how you Inspect Data, Edit and Evolve Neural Networks, all by Graybx:
(no pull requests found since last release)
One platform. Unmatched flexibility. Ultimate control.
Make sure to install our last version on TestPyPI
Happy Training!
What's Changed precisely:
c489fdftrigger CI release in dev
Thank you!
- GuillaumePELLUET
Dev Release v1.4.0.dev0
╭─────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │
│ $$\ $$\ $$\ $$\ $$\ $$\ $$\ │
│ $$ | $\ $$ | \__| $$ | $$ | $$ | $$ | │
│ $$ |$$$\ $$ | $$$$$$\ $$\ $$$$$$\ $$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ │
│ $$ $$ $$\$$ |$$ __$$\ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____|$$ | \____$$\ $$ __$$\ │
│ $$$$ _$$$$ |$$$$$$$$ |$$ |$$ / $$ |$$ | $$ | $$ | \$$$$$$\ $$ | $$$$$$$ |$$ | $$ | │
│ $$$ / \$$$ |$$ ____|$$ |$$ | $$ |$$ | $$ | $$ |$$\ \____$$\ $$ | $$ __$$ |$$ | $$ | │
│ $$ / \$$ |\$$$$$$$\ $$ |\$$$$$$$ |$$ | $$ | \$$$$ |$$$$$$$ |$$$$$$$$\ \$$$$$$$ |$$$$$$$ | │
│ \__/ \__| \_______|\__| \____$$ |\__| \__| \____/ \_______/ \________| \_______|\_______/ │
│ $$\ $$ | │
│ \$$$$$$ | │
│ \______/ │
│ │
│ Inspect - Edit - Evolve Neural Networks │
│ │
╰─── By GrayBx, v1.4.0.dev0 ──────────────────────────────────────────────────────────────────────────╯
Weightslab
Follow us on LinkedIn | Graybx
Welcome
Welcome to the v1.4.0.dev0 release of WeightsLab. This release was made with the idea to improve how you Inspect Data, Edit and Evolve Neural Networks, all by Graybx:
(no pull requests found since last release)
One platform. Unmatched flexibility. Ultimate control.
Make sure to install our last version on TestPyPI
Happy Training!
What's Changed precisely:
72a665dOnly re-run loss-shape autotagging when a signal actually has new data
Thank you!
- GuillaumePELLUET
Dev Release v1.3.4.dev1
╭─────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │
│ $$\ $$\ $$\ $$\ $$\ $$\ $$\ │
│ $$ | $\ $$ | \__| $$ | $$ | $$ | $$ | │
│ $$ |$$$\ $$ | $$$$$$\ $$\ $$$$$$\ $$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ │
│ $$ $$ $$\$$ |$$ __$$\ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____|$$ | \____$$\ $$ __$$\ │
│ $$$$ _$$$$ |$$$$$$$$ |$$ |$$ / $$ |$$ | $$ | $$ | \$$$$$$\ $$ | $$$$$$$ |$$ | $$ | │
│ $$$ / \$$$ |$$ ____|$$ |$$ | $$ |$$ | $$ | $$ |$$\ \____$$\ $$ | $$ __$$ |$$ | $$ | │
│ $$ / \$$ |\$$$$$$$\ $$ |\$$$$$$$ |$$ | $$ | \$$$$ |$$$$$$$ |$$$$$$$$\ \$$$$$$$ |$$$$$$$ | │
│ \__/ \__| \_______|\__| \____$$ |\__| \__| \____/ \_______/ \________| \_______|\_______/ │
│ $$\ $$ | │
│ \$$$$$$ | │
│ \______/ │
│ │
│ Inspect - Edit - Evolve Neural Networks │
│ │
╰─── By GrayBx, v1.3.4.dev1 ──────────────────────────────────────────────────────────────────────────╯
Weightslab
Follow us on LinkedIn | Graybx
Welcome
Welcome to the v1.3.4.dev1 release of WeightsLab. This release was made with the idea to improve how you Inspect Data, Edit and Evolve Neural Networks, all by Graybx:
(no pull requests found since last release)
One platform. Unmatched flexibility. Ultimate control.
Make sure to install our last version on TestPyPI
Happy Training!
What's Changed precisely:
e59bdfaFix doubled "v" in release banner version (vv1.3.4 -> v1.3.4)
Thank you!
- GuillaumePELLUET
Dev Release v1.3.4.dev0
╭─────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │
│ $$\ $$\ $$\ $$\ $$\ $$\ $$\ │
│ $$ | $\ $$ | \__| $$ | $$ | $$ | $$ | │
│ $$ |$$$\ $$ | $$$$$$\ $$\ $$$$$$\ $$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ │
│ $$ $$ $$\$$ |$$ __$$\ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____|$$ | \____$$\ $$ __$$\ │
│ $$$$ _$$$$ |$$$$$$$$ |$$ |$$ / $$ |$$ | $$ | $$ | \$$$$$$\ $$ | $$$$$$$ |$$ | $$ | │
│ $$$ / \$$$ |$$ ____|$$ |$$ | $$ |$$ | $$ | $$ |$$\ \____$$\ $$ | $$ __$$ |$$ | $$ | │
│ $$ / \$$ |\$$$$$$$\ $$ |\$$$$$$$ |$$ | $$ | \$$$$ |$$$$$$$ |$$$$$$$$\ \$$$$$$$ |$$$$$$$ | │
│ \__/ \__| \_______|\__| \____$$ |\__| \__| \____/ \_______/ \________| \_______|\_______/ │
│ $$\ $$ | │
│ \$$$$$$ | │
│ \______/ │
│ │
│ Inspect - Edit - Evolve Neural Networks │
│ │
╰─── By GrayBx, vv1.3.4.dev0 ─────────────────────────────────────────────────────────────────────────╯
Weightslab
Follow us on LinkedIn | Graybx
Welcome
Welcome to the v1.3.4.dev0 release of WeightsLab. This release was made with the idea to improve how you Inspect Data, Edit and Evolve Neural Networks, all by Graybx:
- PR #261: ## What Adds a per-cell loss trajectory to the
GetMetaDatapath and integrates the shape classification with it, in_build_metadata_only_response: - For each visible sample, attach its loss curve as aloss_trajectoryarray DataStat (the on-cell sparkline the studio already renders), viaquery_per_sample(WL_LOSS_TRAJ_SIGNAL, sample_ids). - Classify that same curve withclassify_loss_shapeand emit it as the authoritativetag:loss_shape— so the drawn trajectory and its shap... - PR #260: Rescuing the reactive per-sample signal engine work. This branch (
signal-refinements) was deleted from origin but still held 21 unmerged commits in a local checkout — re-pushing so it's preserved and reviewable. Scope (mostlyweightslab/src.py+backend/logger.py): - Reactive@wl.signalengine — vectorizedbatched=True, multi-input viactx.latest()/inputs=[...], logit-derived signals, in-coreledger_signal_workerthread,StaleSignalErrorfresh-ingest enforcement. - Perf: s... - PR #258: - Add top-level
weightslab launch(and bareweightslab start) to bring up the Studio UI; keepweightslab ui launchworking as a legacy alias. - Auto-open the UI in the default browser after launch (best-effort; skipped in CI/headless/pytest; disable with WEIGHTSLAB_NO_BROWSER). - Update docs + README to the newweightslab launchcommand. - Point "Open in Colab" example links at the renamed wl-*.ipynb notebooks. - Update CLI unit tests for the new launch/start dispatch. - PR #257: This pull request introduces major improvements to the durability and reliability of signal history logging and HDF5 array storage. The logger now supports on-disk persistence via DuckDB, ensuring experiment signal histories are durably stored and checkpointed alongside model weights. The HDF5 array store is made robust against concurrent read/write issues that previously led to file corruption. The old JSON-based logger snapshot system is removed in favor of this new, more efficient approach. **Signal h...
- PR #256: Remove every Docker dependency and include the built UI directly in the Python package. - Before, deps. were Docker + Python - Now, deps. are Python This change includes removing everything related to Docker from code to docs. The UI is built locally from the Weights Studio repository (private), or included in the CI.
- PR #255: Add Tabular examples to WeightsLab: - Fraud Detection - Ads Reco.
- PR #251: Collab compatibility - PyTorch Notebooks - UL Notebooks Integrations Dep issues like NumPy> 2 and protobuf Set min tokens for a model at initialization. Some models need a max token set
One platform. Unmatched flexibility. Ultimate control.
Make sure to install our last version on TestPyPI
Happy Training!
What's Changed precisely:
#261 feat(data_service): per-cell loss trajectory + curve-derived shape (GetMetaData)
e590235feat(data_service): per-cell loss trajectory + curve-derived shaped157c60fix(ci): green the 5 stale dev tests (orphaned bridge + post-#257 log…b6e31ecrefactor(data_service): keep loss-shape classification off the read path206e499refactor(data_service): flatten loss-trajectory block (5 -> 2 nesting…
#260 Rescue: reactive @wl.signal engine + per-sample signal perf (was orphaned branch signal-refinements)
3a4092fexamples: add idiomatic per-sample signals MNIST examplee0421fbexamples: tune ledger_flush_max_rows to cut per-step signal overhead9adfe3efeat(signals): vectorized @wl.signal(batched=True) pathe8b0384feat(signals): multi-input signals via ctx.latest() / bctx.latest()d0605b8feat(signals): enforce fresh ingests (StaleSignalError), stop swallow…ea3e9eafeat(signals): reactive inputs=[...] — order-independent multi-input …093e6c4feat(signals): in-core signal-worker thread (ledger_signal_worker)b8e3492perf(signals): step-scoped query cache + value-at-step gather; cycle …4731598perf(signals): staging-buffer value-at-step read + vectorized id->sca…2858b72feat(signals): logit-derived signals + batched reactive persist + pre…090ab94examples: rewrite ws-signals-mnist to zero save_signals (ctx.logits +…4a6871dperf(signals): don't query the ledger for reactive-derived inputs mid…bb11318examples: unify signals-mnist with the stress harness (train+eval, li…3ccc726tests: unit tests for signal-refinements (query cache, value-at-step,…3cb6e19Fix merged conflicts1414e0cupdate gitignore - remove coverage filesfea2758remove last merged conflicts5e0ed70logger: make query-cache maxsize env-configurable (WL_QUERY_CACHE_MAX…b3f09ebfeat(signals): default loss-shape classifier + report summary (PR #24…c8610f7refactor(signals): clean, layered, overridable shape API
#258 Rename UI launch command; auto-open browser; update Colab links to wl-*
bd8a48bRename UI launch command; auto-open browser; update Colab links to wl-*4cd67f5fix merged conflicts
#257 H5 and Loggers fixes
628a89bFix bug with h5 filecaa29ddChange from old logger dumping behavior (jsonz) to DB dump
#256 Docker Dependency Removal
7f3874cAdd Docker-free pip-native UI:weightslab startf45510bRemove Docker from weightslab; rename CLI to weightslab.cli; Docker-f…
#255 Add two tabular PyTorch examples: bank fraud detection + ads CTR reco…
d11f077Add two tabular PyTorch examples: bank fraud detection + ads CTR reco…15876f2Make tabular examples fully WeightsLab-native (feature columns) + ver…538362dSupport tabular (1-D feature vector) inputs through the gRPC data patheca156ddocs: examples gallery + Colab notebooks for the tabular fraud & ads …
#251 v1.3.4.dev0 Full Collab Integration and bug fixed
c4bb3dbv1.3.3: Release — Collab integration, UI improvements, CLI enhancemen…c0b1953Update CHANGELOGf667f7bSet templates READ ONLY collabce908c6fix utests and change other thingsdc75438lower pandas floore85f646Fix protobuf dep versionb7f089bFix export function in json to default orient columnaraecbc26Fix the CLS issue when stopping a cell3ce3d01Perf: halve view-rebuild cost by removing duplicate get_combined_df pull32cc4d7Notebook(cls): move dataloader params into nested config["data"][loader]97ff269Notebook(cls): detach bore tunnel so cell-stop SIGINT doesn't kill it4ec1279Fix write_history JSON crash; align write_dataframe tests with column…c23ebf3Fix notebooks4218f67Bore freedom, no parent process if notebookbfd0aa6Fix bugs, config parameters, and run2371217Notebook(cls): add section 10 — curate in the UI to boost performance637aaacNotebook(seg): load dataset from a Google Drive zip instead of repo s…515cbdfAdd gdrive dataset to train BDD herec7d8393Notebook(seg): fix BCE shape mismatch — resize image+mask to a fi...
Dev Release v1.3.3.dev7
╭─────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │
│ $$\ $$\ $$\ $$\ $$\ $$\ $$\ │
│ $$ | $\ $$ | \__| $$ | $$ | $$ | $$ | │
│ $$ |$$$\ $$ | $$$$$$\ $$\ $$$$$$\ $$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ │
│ $$ $$ $$\$$ |$$ __$$\ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____|$$ | \____$$\ $$ __$$\ │
│ $$$$ _$$$$ |$$$$$$$$ |$$ |$$ / $$ |$$ | $$ | $$ | \$$$$$$\ $$ | $$$$$$$ |$$ | $$ | │
│ $$$ / \$$$ |$$ ____|$$ |$$ | $$ |$$ | $$ | $$ |$$\ \____$$\ $$ | $$ __$$ |$$ | $$ | │
│ $$ / \$$ |\$$$$$$$\ $$ |\$$$$$$$ |$$ | $$ | \$$$$ |$$$$$$$ |$$$$$$$$\ \$$$$$$$ |$$$$$$$ | │
│ \__/ \__| \_______|\__| \____$$ |\__| \__| \____/ \_______/ \________| \_______|\_______/ │
│ $$\ $$ | │
│ \$$$$$$ | │
│ \______/ │
│ │
│ Inspect - Edit - Evolve Neural Networks │
│ │
╰─── By GrayBx, vv1.3.3.dev7 ─────────────────────────────────────────────────────────────────────────╯
Weightslab
Follow us on LinkedIn | Graybx
Welcome
Welcome to the v1.3.3.dev7 release of WeightsLab. This release was made with the idea to improve how you Inspect Data, Edit and Evolve Neural Networks, all by Graybx:
(no pull requests found since last release)
One platform. Unmatched flexibility. Ultimate control.
Make sure to install our last version on TestPyPI
Happy Training!
What's Changed precisely:
172a8ecfix missing import61e8292fix AspectRatiod901a4dNotebook(seg): convert to sample-wise only (fix BCE shape crash, drop per-instance)c7d8393Notebook(seg): fix BCE shape mismatch — resize image+mask to a fixed square515cbdfAdd gdrive dataset to train BDD here637aaacNotebook(seg): load dataset from a Google Drive zip instead of repo sparse-checkout2371217Notebook(cls): add section 10 — curate in the UI to boost performancebfd0aa6Fix bugs, config parameters, and run
Thank you!
- Guillaume
- GuillaumePELLUET
Dev Release v1.3.3.dev6
╭─────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │
│ $$\ $$\ $$\ $$\ $$\ $$\ $$\ │
│ $$ | $\ $$ | \__| $$ | $$ | $$ | $$ | │
│ $$ |$$$\ $$ | $$$$$$\ $$\ $$$$$$\ $$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ │
│ $$ $$ $$\$$ |$$ __$$\ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____|$$ | \____$$\ $$ __$$\ │
│ $$$$ _$$$$ |$$$$$$$$ |$$ |$$ / $$ |$$ | $$ | $$ | \$$$$$$\ $$ | $$$$$$$ |$$ | $$ | │
│ $$$ / \$$$ |$$ ____|$$ |$$ | $$ |$$ | $$ | $$ |$$\ \____$$\ $$ | $$ __$$ |$$ | $$ | │
│ $$ / \$$ |\$$$$$$$\ $$ |\$$$$$$$ |$$ | $$ | \$$$$ |$$$$$$$ |$$$$$$$$\ \$$$$$$$ |$$$$$$$ | │
│ \__/ \__| \_______|\__| \____$$ |\__| \__| \____/ \_______/ \________| \_______|\_______/ │
│ $$\ $$ | │
│ \$$$$$$ | │
│ \______/ │
│ │
│ Inspect - Edit - Evolve Neural Networks │
│ │
╰─── By GrayBx, vv1.3.3.dev6 ─────────────────────────────────────────────────────────────────────────╯
Weightslab
Follow us on LinkedIn | Graybx
Welcome
Welcome to the v1.3.3.dev6 release of WeightsLab. This release was made with the idea to improve how you Inspect Data, Edit and Evolve Neural Networks, all by Graybx:
(no pull requests found since last release)
One platform. Unmatched flexibility. Ultimate control.
Make sure to install our last version on TestPyPI
Happy Training!
What's Changed precisely:
4218f67Bore freedom, no parent process if notebook
Thank you!
- GuillaumePELLUET
Dev Release v1.3.3.dev5
╭─────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │
│ $$\ $$\ $$\ $$\ $$\ $$\ $$\ │
│ $$ | $\ $$ | \__| $$ | $$ | $$ | $$ | │
│ $$ |$$$\ $$ | $$$$$$\ $$\ $$$$$$\ $$$$$$$\ $$$$$$\ $$$$$$$\ $$ | $$$$$$\ $$$$$$$\ │
│ $$ $$ $$\$$ |$$ __$$\ $$ |$$ __$$\ $$ __$$\ \_$$ _| $$ _____|$$ | \____$$\ $$ __$$\ │
│ $$$$ _$$$$ |$$$$$$$$ |$$ |$$ / $$ |$$ | $$ | $$ | \$$$$$$\ $$ | $$$$$$$ |$$ | $$ | │
│ $$$ / \$$$ |$$ ____|$$ |$$ | $$ |$$ | $$ | $$ |$$\ \____$$\ $$ | $$ __$$ |$$ | $$ | │
│ $$ / \$$ |\$$$$$$$\ $$ |\$$$$$$$ |$$ | $$ | \$$$$ |$$$$$$$ |$$$$$$$$\ \$$$$$$$ |$$$$$$$ | │
│ \__/ \__| \_______|\__| \____$$ |\__| \__| \____/ \_______/ \________| \_______|\_______/ │
│ $$\ $$ | │
│ \$$$$$$ | │
│ \______/ │
│ │
│ Inspect - Edit - Evolve Neural Networks │
│ │
╰─── By GrayBx, vv1.3.3.dev5 ─────────────────────────────────────────────────────────────────────────╯
Weightslab
Follow us on LinkedIn | Graybx
Welcome
Welcome to the v1.3.3.dev5 release of WeightsLab. This release was made with the idea to improve how you Inspect Data, Edit and Evolve Neural Networks, all by Graybx:
(no pull requests found since last release)
One platform. Unmatched flexibility. Ultimate control.
Make sure to install our last version on TestPyPI
Happy Training!
What's Changed precisely:
c23ebf3Fix notebooks4ec1279Fix write_history JSON crash; align write_dataframe tests with columnar export
Thank you!
- GuillaumePELLUET