All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
- PR #1729: Add support for Python 3.11 and remove support for all other minor versions
Thanks to @bhancock8, @kamelCased, @MLobo1997, and @rsmith49 for contributions!
- PR #1690: Bump numpy version for Mac M1 compat
- PR #1696: Fix linting
- PR #1694: Fix test loss
- PR #1693: Update black version to fix build
- PR #1671: Fix flaky test
- PR #1688: Fix branch filtering for complex tests
- PR #1686: Switch to CircleCI badge
- PR #1685: Migrating travis -> circle
Thanks to @rsmith49, @zexuan-zhou, @humzaiqbal, @fpoms, @crawlingcub, and @henryre for contributions!
- PR #1649: Add progress bar for label model training
- PR #1645: Upgrade networkx requirement to support networkx>=2.5
- PR #1663: Upgrade tensorboard requirement to support tensorboard>=2.0
- PR #1652: Fix issue with logging display during label model training
- PR #1677: Bump version upper bound of networkx to 2.7
Thanks to @anerirana, @thatch, @sfilipi, @hardianlawi, @asottile and @marekmodry for contributions!
- PR #1633, #1628: Update requirements for
numpy
(>=1.16.5,<1.20.0),pandas
(>=1.0.0,<2.0.0), andscikit-learn
(>=0.20.2,<0.25.0)
- PR #1628: Add unit test coverage for Python 3.8
- PR #1616: Accept prec_init as array or list in
LabelModel
- PR #1602: Add get_label_instances to analysis module
Thanks to @DavidKoleczek and @antonis19 for contributions!
- PR #1572: Allow specification of
memoize_key
in preprocessors - PR #1597: Improved error messages for MultitaskClassifier
- PR #1592: Improved LabelModel documentation
Thanks to @Wirg, @TrigonaMinima, and @dchichkov for recent contributions!
- PR #1565: Upgrade pytorch requirement to support torch>=1.2.0
- PR #1535: Refactor baseline model imports in snorkel.labeling.model
- Now, from
from snorkel.labeling import MajorityLabelVoter, LabelModel
can be expressedfrom snorkel.labeling.model import MajorityLabelVoter, LabelModel
- Now, from
- PR #1533: Allow option to save optimizer state in Trainer
- PR #1523: Add GPU option for spaCy preprocessor
- PR #1540: Fix squeeze bug in
to_int_label_array
function - PR #1520: Fix error bucket documentation in snorkel.analysis.error_analysis
Thanks to @rjurney, @ptrcklv for recent contributions!
- PR #1502: Faster symmetry breaking in LabelModel using Munkres algorithm
- PR #1481: removed fault tolerant mode for labeling functions
- PR #1481: fault tolerant mode for appliers
- PR #1450, 1467: ignore abstains in scoring, except coverage
- PR #1463: serialize all attributes of label model
- PR #1466: fix label model GPU training option
- PR #1477, #1492: pin dependency versions
- PR #1454:
set_seed
utility removed
Thanks to @HiromuHota, @ferhatelmas, and @garaud for their recent contributions!
- PR #1453:
SlicingClassifier
renamed toSliceAwareClassifier
- PR #1451: add heuristic for breaking symmetry in multiple label model optima case
- PR #1442: integration test for
MultitaskClassifier
- PR #1444: fix label model weight clamping behavior
- PR #1445: fix JSON log writer
- PR #1447: fix correct/incorrect count bug in
LFAnalysis
- PR #1428, 1449: catch invalid label model inputs
- PR #1441: make inputs to
Scorer.score
optional
Version 0.9.0 is a complete redesign of the Snorkel library. There's too much added, changed, and removed to list in this entry. For more information on the release, check out the project homepage. From here forward, we'll keep a detailed changelog.