v0.6.4

v0.6.4 (2024-05-13)

Build

• build: remove upper pin on hypothesis and torch range (bfe6953)

Chore

• chore(deps): bump ruff from 0.4.3 to 0.4.4

Bumps ruff from 0.4.3 to 0.4.4.

• Release notes
• Changelog
• Commits

---

updated-dependencies:

• dependency-name: ruff
  dependency-type: direct:production
  update-type: version-update:semver-patch
  ...

Signed-off-by: dependabot[bot] <[email protected]> (aa33cf8)

• chore(deps): bump mkdocs-material from 9.4.6 to 9.5.22

Bumps mkdocs-material from 9.4.6 to 9.5.22.

• Release notes
• Changelog
• Commits

---

updated-dependencies:

• dependency-name: mkdocs-material
  dependency-type: direct:production
  update-type: version-update:semver-minor
  ...

Signed-off-by: dependabot[bot] <[email protected]> (34558be)

• chore(deps): update transformers requirement

Updates the requirements on transformers to permit the latest version.

• Release notes
• Commits

---

updated-dependencies:

• dependency-name: transformers
  dependency-type: direct:production
  ...

Signed-off-by: dependabot[bot] <[email protected]> (06102ea)

• chore(deps): bump dawidd6/action-download-artifact from 2.24.3 to 3.1.4

Bumps dawidd6/action-download-artifact from 2.24.3 to 3.1.4.

• Release notes
• Commits

---

updated-dependencies:

• dependency-name: dawidd6/action-download-artifact
  dependency-type: direct:production
  update-type: version-update:semver-major
  ...

Signed-off-by: dependabot[bot] <[email protected]> (48bc9e3)

Ci

• ci: increase dependabot frequency (e44cd5f)

• ci: fix missing coverage comment by creating pytest-coverage.txt file during build (a8772c9)

• ci: store hypothesis database in GitHub Actions to allow reproducing locally (fd62aca)

Unknown

• Merge pull request #39 from qthequartermasterman/dependabot/github_actions/main/dawidd6/action-download-artifact-3.1.4

chore(deps): bump dawidd6/action-download-artifact from 2.24.3 to 3.1.4 (a83b029)

• Merge pull request #43 from qthequartermasterman/dependabot/pip/main/mkdocs-material-9.5.22

chore(deps): bump mkdocs-material from 9.4.6 to 9.5.22 (2f10d4c)

• Merge pull request #45 from qthequartermasterman/remove-upper-dependency-pin

build: remove upper pin on hypothesis and torch range (0d0d472)

• Merge pull request #44 from qthequartermasterman/dependabot/pip/main/ruff-0.4.4

chore(deps): bump ruff from 0.4.3 to 0.4.4 (8513f17)

• Merge pull request #41 from qthequartermasterman/dependabot/pip/main/transformers-gte-4.0.0-and-lte-4.40.2

chore(deps): update transformers requirement from <=4.40.0,>=4.0.0 to >=4.0.0,<=4.40.2 (7e9a01a)

• Merge pull request #38 from qthequartermasterman/dependabot-update

ci: increase dependabot frequency (229887c)

• Merge pull request #37 from qthequartermasterman/fix-coverage-txt

ci: fix missing coverage comment by creating pytest-coverage.txt file during build (56eb109)

• Merge pull request #36 from qthequartermasterman/github-actions-database

ci: store hypothesis database in GitHub Actions to allow reproducing failures locally (0a2eb72)

v0.6.3

v0.6.3 (2024-05-12)

Documentation

• docs: update pyproject toml classifiers, keywords, and urls. (e6d4a7c)

• docs(badges): 📝 Add pypi badges for downloads and version to readme (00563e6)

• docs(version): 📝 remove reference to python 3.8 in classifiers (03322e1)

Fix

• fix: 🐛 fix incompatible parameters when generating Adam optimizers on Torch 1.13.

Fix incompatible parameters:

• fused is only available if cuda is available
• fused is incompatible with differentiable (d0ef6bb)

Unknown

• Merge pull request #35 from qthequartermasterman/update-pyproject-toml

docs: update pyproject toml classifiers, keywords, and urls. (0d3a3ab)

• Merge pull request #34 from CandiedCode/badges (f166a9e)

• Merge pull request #33 from CandiedCode/version (f9ff580)

v0.6.2

v0.6.2 (2024-05-10)

Build

• build: 🐛 set dependency ranges for optimal compatability (4a25c1d)

Unknown

• Merge pull request #32 from qthequartermasterman/set-dependencies-ranges

build: 🐛 set dependency ranges for optimal compatibility (71c52b7)

v0.6.1

v0.6.1 (2024-05-10)

Build

• build: ⚡ remove unnecessary tee to avoid supressing error code (8716033)

• build: fix incorrect reusable workflow path (ec36fc6)

• build: test pytorch 1.13 in CI (6aff502)

• build: convert workflow to reusable workflow for simplicity (b1aa17a)

• build: ✨ 📝 Add readthedocs configuration. (95ffac5)

Documentation

• docs: ✨ 📝 Add index page (510e3df)

• docs: ✨ 📝 Add strategies reference to docs. (8fc0d95)

• docs: ✨ 📝 Add compatability guide to docs. (ca3bada)

• docs: ✨ 📝 Disable top tabs on docs. (53164a2)

• docs: ✨ 📝 Add quick start guide to docs. (2710504)

• docs: ✨ 📝 Add quick start guide to docs. (a59871c)

• docs: ✨ 📝 Add contribuging guidelines to docs. (3f30767)

• docs: ✨ 📝 Add changelog to docs. (05dc054)

• docs: ✨ 📝 Add changelog to docs. (ddb9598)

• docs: ✨ 📝 Add license file to docs. (06adb42)

• docs: ✨ Initial docsite set up. (274bb65)

Fix

• fix: 🐛 do not try to use the device as a context manager in torch<2. (3114162)

• fix: 🐛 do not try to use the meta device as a context manager in torch<2. (ef0e47c)

• fix: ⚡ remove overly-restrictive hypothesis deadline for transformers tests (d1c89f2)

Test

• test: 🐛 ignore runtime errors during module import (882f515)

• test: 🐛 ignore runtime errors during module import (5c89c79)

Unknown

• Merge pull request #31 from qthequartermasterman/test-multiple-torch-versions

build: Test multiple torch versions (4ecabd9)

• Merge pull request #29 from qthequartermasterman/docsite

docs: ✨ 📝 Add index page (46e3516)

• Merge pull request #28 from qthequartermasterman/docsite

docs: Add support for docs rendered by readthedocs, powered by mkdocs. (e8de564)

v0.6.0

v0.6.0 (2024-05-09)

Unknown

• Merge pull request #18 from qthequartermasterman/optimizers

feat: ✨ add strategies for torch optimizers (fbefa06)

• Merge branch 'main' into optimizers (4eb2f4d)

v0.5.0

v0.5.0 (2024-05-09)

Build

• build: 🔧 add py.typed (bef3985)

Chore

• chore(deps): bump pytest from 8.1.1 to 8.2.0

Bumps pytest from 8.1.1 to 8.2.0.

• Release notes
• Changelog
• Commits

---

updated-dependencies:

• dependency-name: pytest
  dependency-type: direct:production
  update-type: version-update:semver-minor
  ...

Signed-off-by: dependabot[bot] <[email protected]> (75fd127)

• chore(deps): bump ruff from 0.4.1 to 0.4.3

Bumps ruff from 0.4.1 to 0.4.3.

• Release notes
• Changelog
• Commits

---

updated-dependencies:

• dependency-name: ruff
  dependency-type: direct:production
  update-type: version-update:semver-patch
  ...

Signed-off-by: dependabot[bot] <[email protected]> (c7e5d8d)

Fix

• fix: 🏷️ fix incorrect decorator typing (3e80d6a)

• fix: ✅ designate officially supported versus unofficially supported transformers.

By default, only officially supported transformer types are tested in CI, but unofficially supported transformers MAY or MAY NOT be compatible with the transformer_strategy. A test fully parametrized over ALL of the model types in transformers is available, and can be enabled when setting the environment variable HYPOTHESIS_TORCH_TEST_UNSUPPORTED_TRANSFORMERS=True. (7d22190)

Test

• test: ignore import errors while inferring available transformers in tests. (84185cb)

• test: disable deadline for occasionally slow tests (34ae9db)

Unknown

• Merge pull request #19 from qthequartermasterman/transformers-tests

feat: ✨ add support for more transformers strategies (964a70f)

• Merge branch 'main' into transformers-tests (aa6f06e)

• Merge pull request #23 from qthequartermasterman/dependabot/pip/main/pytest-8.2.0

chore(deps): bump pytest from 8.1.1 to 8.2.0 (a55308b)

• Merge pull request #26 from qthequartermasterman/dependabot/pip/main/ruff-0.4.3

chore(deps): bump ruff from 0.4.1 to 0.4.3 (2ae1ae5)

• Merge pull request #25 from ringohoffman/py.typed

build: 🔧 add py.typed (b2e8557)

v0.4.3

v0.4.3 (2024-04-28)

Fix

• fix: 🐛 do not generate inf values if elements has infinity disabled for bfloat16 tensors.

Before, if a float strategy that prevented infinities was provided to as elements to tensor_strategy, the resultant tensor could still occasionally have infinities because the strategy would generate float32 values that exceeded the maximium/minimum values for bfloat16. (cda8577)

Unknown

• Merge pull request #21 from qthequartermasterman/fix-bfloat16-infinities

fix: 🐛 do not generate inf values if elements has infinity disabled for bfloat16 tensors. (8398018)

v0.4.2

v0.4.2 (2024-04-27)

Fix

• fix: 🐛 do not generate inf values if elements has infinity disabled for float16 and float32 tensors.

This fix does not work for all bfloat16 tensors. (8e9eb5d)

v0.4.1

v0.4.1 (2024-04-24)

Fix

• fix: mark bfloat16 as a floating dtype (38b4ed3)

Unknown

• Merge pull request #17 from qthequartermasterman/register-random-state

fix: mark bfloat16 as a floating dtype (57c0a5b)

v0.4.0

v0.4.0 (2024-04-24)

Feature

• feat: register random torch state as a hypothesis entry point (so that torch state is always registered, even if hypothesis-torch is never imported). (eada00a)

• feat: register random torch state with hypothesis (633bf82)

Unknown

• Merge pull request #16 from qthequartermasterman/register-random-state

feat: Register PyTorch's random state with Hypothesis so that torch random is deterministic during testing (4d957a1)

• Update issue templates (c7648bf)