axel7083
commented
2 months ago llamacpp python ==0.2.82 -> ==0.2.84 introduce a regression when using the LlamaGrammar. I think it is better to wait before merging this bump
see https://github.com/abetlen/llama-cpp-python/issues/1623 and https://github.com/abetlen/llama-cpp-python/issues/1636
rhatdan
This PR contains the following updates:
==1.24.7->==1.24.9==0.5.4->==0.5.5==0.0.4->==0.0.5==0.2.82->==0.2.85==1.0.7->==1.0.8==4.66.4->==4.66.5==0.30.3->==0.30.5Release Notes
pymupdf/pymupdf (PyMuPDF)
### [`v1.24.9`](https://togithub.com/pymupdf/PyMuPDF/releases/tag/1.24.9): PyMuPDF-1.24.9 released [Compare Source](https://togithub.com/pymupdf/pymupdf/compare/1.24.8...1.24.9) PyMuPDF-1.24.9 has been released. Wheels for Windows, Linux and MacOS, and the sdist, are available on pypi.org and can be installed in the usual way, for example: python -m pip install --upgrade pymupdf \[Linux-aarch64 wheels will be built and uploaded later.] **Changes in version 1.24.9 (2024-07-24)** - Incremented MuPDF version to 1.24.8. ### [`v1.24.8`](https://togithub.com/pymupdf/PyMuPDF/releases/tag/1.24.8): PyMuPDF-1.24.8 released [Compare Source](https://togithub.com/pymupdf/pymupdf/compare/1.24.7...1.24.8) PyMuPDF-1.24.8 has been released. Wheels for Windows, Linux and MacOS, and the sdist, are available on pypi.org and can be installed in the usual way, for example: python -m pip install --upgrade pymupdf \[Linux-aarch64 wheels will be built and uploaded later.] **Changes in version 1.24.8 (2024-07-22)** - Fixed issues: - **Fixed** [#3636](https://togithub.com/pymupdf/pymupdf/issues/3636) - **Fixed** [#3654](https://togithub.com/pymupdf/pymupdf/issues/3654) - **Fixed** [#3677](https://togithub.com/pymupdf/pymupdf/issues/3677) - **Fixed** [#3687](https://togithub.com/pymupdf/pymupdf/issues/3687) Other: - Fixed various spelling mistakes spotted by codespell. - Improved how we modify MuPDF's default configuration on Windows. - Make text search to work with ligatures.chroma-core/chroma (chromadb)
### [`v0.5.5`](https://togithub.com/chroma-core/chroma/releases/tag/0.5.5) [Compare Source](https://togithub.com/chroma-core/chroma/compare/0.5.4...0.5.5) Version: `0.5.5` Git ref: `refs/tags/0.5.5` Build Date: `2024-07-23T01:01` PIP Package: `chroma-0.5.5.tar.gz` Github Container Registry Image: `ghcr.io/chroma-core/chroma:0.5.5` DockerHub Image: `chromadb/chroma:0.5.5` #### What's Changed - \[PERF] Better tracing for storage access by [@HammadB](https://togithub.com/HammadB) in [https://github.com/chroma-core/chroma/pull/2500](https://togithub.com/chroma-core/chroma/pull/2500) - \[BUG] Fix cross version persistence tests after 0.5.4 release by [@HammadB](https://togithub.com/HammadB) in [https://github.com/chroma-core/chroma/pull/2501](https://togithub.com/chroma-core/chroma/pull/2501) - \[BLD] reduce max number of examples on slow prop test preset by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2504](https://togithub.com/chroma-core/chroma/pull/2504) - \[PERF] Trace block read size by [@HammadB](https://togithub.com/HammadB) in [https://github.com/chroma-core/chroma/pull/2505](https://togithub.com/chroma-core/chroma/pull/2505) - \[ENH] Binary search for blockfile get() by [@sanketkedia](https://togithub.com/sanketkedia) in [https://github.com/chroma-core/chroma/pull/2506](https://togithub.com/chroma-core/chroma/pull/2506) - \[ENH] Debuggable test_sanity by [@sanketkedia](https://togithub.com/sanketkedia) in [https://github.com/chroma-core/chroma/pull/2507](https://togithub.com/chroma-core/chroma/pull/2507) - \[PERF] Materialization traces by [@HammadB](https://togithub.com/HammadB) in [https://github.com/chroma-core/chroma/pull/2508](https://togithub.com/chroma-core/chroma/pull/2508) - \[ENH] Add collection id as log entry in test_embeddings by [@sanketkedia](https://togithub.com/sanketkedia) in [https://github.com/chroma-core/chroma/pull/2472](https://togithub.com/chroma-core/chroma/pull/2472) - \[BUG] fix issue when deleting->adding same ID before index is persisted by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2512](https://togithub.com/chroma-core/chroma/pull/2512) - \[BUG] fix persistent HNSW parameter migration by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2511](https://togithub.com/chroma-core/chroma/pull/2511) - \[ENH] updates & invalid operations should also trigger persisting of local HNSW by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2499](https://togithub.com/chroma-core/chroma/pull/2499) - \[BUG] fix invalid operation persist check by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2520](https://togithub.com/chroma-core/chroma/pull/2520) - \[ENH] delete collection during teardown of state machine by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2521](https://togithub.com/chroma-core/chroma/pull/2521) - \[ENH] Use foyer for block cache by [@Ishiihara](https://togithub.com/Ishiihara) in [https://github.com/chroma-core/chroma/pull/2431](https://togithub.com/chroma-core/chroma/pull/2431) - \[ENH] Introduce stream abstraction and enable concurrency test for blockfile by [@Ishiihara](https://togithub.com/Ishiihara) in [https://github.com/chroma-core/chroma/pull/2454](https://togithub.com/chroma-core/chroma/pull/2454) - \[ENH] Allow cache eviction for HNSW provider by [@Ishiihara](https://togithub.com/Ishiihara) in [https://github.com/chroma-core/chroma/pull/2484](https://togithub.com/chroma-core/chroma/pull/2484) - \[BUG] CHROMA_SERVER_NOFILE cannot be overriden with env var in container by [@tazarov](https://togithub.com/tazarov) in [https://github.com/chroma-core/chroma/pull/2209](https://togithub.com/chroma-core/chroma/pull/2209) - \[ENH] Use instrumented tracing for HNSW provider file read by [@Ishiihara](https://togithub.com/Ishiihara) in [https://github.com/chroma-core/chroma/pull/2525](https://togithub.com/chroma-core/chroma/pull/2525) - \[BLD] Allow query/compactor to specify tolerations in k8s by [@HammadB](https://togithub.com/HammadB) in [https://github.com/chroma-core/chroma/pull/2527](https://togithub.com/chroma-core/chroma/pull/2527) - \[RELEASE] Release helm 0.1.4 by [@HammadB](https://togithub.com/HammadB) in [https://github.com/chroma-core/chroma/pull/2530](https://togithub.com/chroma-core/chroma/pull/2530) - \[CLN] Make LocalStorage use sync filesystem APIs by [@Ishiihara](https://togithub.com/Ishiihara) in [https://github.com/chroma-core/chroma/pull/2531](https://togithub.com/chroma-core/chroma/pull/2531) - \[BUG] fix HNSW param defaults in new configuration logic & require batch_size < sync_threshold by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2526](https://togithub.com/chroma-core/chroma/pull/2526) - \[BUG]: fix missing record in query result when many records were deleted and pending persist by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2532](https://togithub.com/chroma-core/chroma/pull/2532) - \[CLN] Move max_block_size_bytes to block_manager_config by [@Ishiihara](https://togithub.com/Ishiihara) in [https://github.com/chroma-core/chroma/pull/2535](https://togithub.com/chroma-core/chroma/pull/2535) - \[TST] Get ts-integration-test.sh script working locally by [@AlabasterAxe](https://togithub.com/AlabasterAxe) in [https://github.com/chroma-core/chroma/pull/2518](https://togithub.com/chroma-core/chroma/pull/2518) - \[ENH] simplify logic for when to persist index changes by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2539](https://togithub.com/chroma-core/chroma/pull/2539) - \[ENH] Add an init method that eagerly reports errors with the tenant or DB by [@AlabasterAxe](https://togithub.com/AlabasterAxe) in [https://github.com/chroma-core/chroma/pull/2537](https://togithub.com/chroma-core/chroma/pull/2537) - \[BUG] debug nightly tests timing out by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2538](https://togithub.com/chroma-core/chroma/pull/2538) - Revert "\[ENH] simplify logic for when to persist index changes" by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2544](https://togithub.com/chroma-core/chroma/pull/2544) - \[TST] Enable cache eviction for shuttle concurrency test by [@Ishiihara](https://togithub.com/Ishiihara) in [https://github.com/chroma-core/chroma/pull/2546](https://togithub.com/chroma-core/chroma/pull/2546) - \[ENH] Add error trace in log service for GetAllCollectionInfoToCompact by [@sanketkedia](https://togithub.com/sanketkedia) in [https://github.com/chroma-core/chroma/pull/2548](https://togithub.com/chroma-core/chroma/pull/2548) - \[BLD] log artifacts per python version by [@sanketkedia](https://togithub.com/sanketkedia) in [https://github.com/chroma-core/chroma/pull/2556](https://togithub.com/chroma-core/chroma/pull/2556) - \[CHORE] bump hnswlib by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2558](https://togithub.com/chroma-core/chroma/pull/2558) - \[RELEASE] 0.5.5 by [@codetheweb](https://togithub.com/codetheweb) in [https://github.com/chroma-core/chroma/pull/2559](https://togithub.com/chroma-core/chroma/pull/2559) **Full Changelog**: https://github.com/chroma-core/chroma/compare/0.5.4...0.5.5fastapi/fastapi-cli (fastapi-cli)
### [`v0.0.5`](https://togithub.com/fastapi/fastapi-cli/blob/HEAD/release-notes.md#005) [Compare Source](https://togithub.com/fastapi/fastapi-cli/compare/0.0.4...0.0.5) ##### Breaking Changes - ♻️ Add `fastapi-cli[standard]` including Uvicorn, make `fastapi-cli` and `fastapi-cli-slim` have the same packages. PR [#55](https://togithub.com/fastapi/fastapi-cli/pull/55) by [@tiangolo](https://togithub.com/tiangolo). - ➕ Keep Uvicorn in default dependencies. PR [#57](https://togithub.com/fastapi/fastapi-cli/pull/57) by [@tiangolo](https://togithub.com/tiangolo). ##### Summary Install with: ```bash pip install "fastapi[standard]" ``` Or if for some reason installing only the FastAPI CLI: ```bash pip install "fastapi-cli[standard]" ``` ##### Technical Details Before this, `fastapi-cli` would include Uvicorn and `fastapi-cli-slim` would not include Uvicorn. In a future version, `fastapi-cli` will not include Uvicorn unless it is installed with `fastapi-cli[standard]`. FastAPI version 0.112.0 has a `fastapi[standard]` and that one includes `fastapi-cli[standard]`. Before, you would install `pip install fastapi`, or `pip install fastapi-cli`. Now you should include the `standard` optional dependencies (unless you want to exclude one of those): `pip install "fastapi[standard]"`. In a future version, `fastapi-cli` will not include Uvicorn unless it is installed with `fastapi-cli[standard]`. ##### Refactors - ♻️ Simplify code in `src/fastapi_cli/discover.py`. PR [#22](https://togithub.com/tiangolo/fastapi-cli/pull/22) by [@pedroimpulcetto](https://togithub.com/pedroimpulcetto). ##### Docs - 🚚 Rename repo references to new GitHub FastAPI org. PR [#56](https://togithub.com/fastapi/fastapi-cli/pull/56) by [@tiangolo](https://togithub.com/tiangolo). ##### Internal - ⬆ Bump ruff from 0.4.5 to 0.5.5. PR [#52](https://togithub.com/fastapi/fastapi-cli/pull/52) by [@dependabot\[bot\]](https://togithub.com/apps/dependabot). - 🔧 Remove Python 3.7 from PyPI classifiers. PR [#48](https://togithub.com/fastapi/fastapi-cli/pull/48) by [@patrick91](https://togithub.com/patrick91). - ⬆ \[pre-commit.ci] pre-commit autoupdate. PR [#28](https://togithub.com/fastapi/fastapi-cli/pull/28) by [@pre-commit-ci\[bot\]](https://togithub.com/apps/pre-commit-ci). - ⬆ Bump mypy from 1.10.0 to 1.11.1. PR [#53](https://togithub.com/fastapi/fastapi-cli/pull/53) by [@dependabot\[bot\]](https://togithub.com/apps/dependabot). - ⬆ Bump pypa/gh-action-pypi-publish from 1.8.14 to 1.9.0. PR [#34](https://togithub.com/fastapi/fastapi-cli/pull/34) by [@dependabot\[bot\]](https://togithub.com/apps/dependabot). - 👷 Update issue-manager.yml GitHub Action permissions. PR [#54](https://togithub.com/tiangolo/fastapi-cli/pull/54) by [@tiangolo](https://togithub.com/tiangolo). - ⬆ Bump ruff from 0.4.4 to 0.4.5. PR [#29](https://togithub.com/tiangolo/fastapi-cli/pull/29) by [@dependabot\[bot\]](https://togithub.com/apps/dependabot). - ⬆ Bump ruff from 0.4.3 to 0.4.4. PR [#23](https://togithub.com/tiangolo/fastapi-cli/pull/23) by [@dependabot\[bot\]](https://togithub.com/apps/dependabot). - 👷 Enable CI tests for Python 3.12. PR [#27](https://togithub.com/tiangolo/fastapi-cli/pull/27) by [@tiangolo](https://togithub.com/tiangolo). - 👷 Update Upload/Download artifacts GitHub Actions. PR [#26](https://togithub.com/tiangolo/fastapi-cli/pull/26) by [@tiangolo](https://togithub.com/tiangolo).abetlen/llama-cpp-python (llama-cpp-python)
### [`v0.2.85`](https://togithub.com/abetlen/llama-cpp-python/blob/HEAD/CHANGELOG.md#0285) [Compare Source](https://togithub.com/abetlen/llama-cpp-python/compare/v0.2.84...v0.2.85) - feat: Update llama.cpp to [ggerganov/llama.cpp@`398ede5`](https://togithub.com/ggerganov/llama.cpp/commit/398ede5efeb07b9adf9fbda7ea63f630d476a792) - fix: Missing LoRA adapter after API change by [@shamitv](https://togithub.com/shamitv) in [#1630](https://togithub.com/abetlen/llama-cpp-python/issues/1630) - fix(docker): Update Dockerfile BLAS options by [@olivierdebauche](https://togithub.com/olivierdebauche) in [#1632](https://togithub.com/abetlen/llama-cpp-python/issues/1632) - fix(docker): Fix GGML_CUDA param by [@olivierdebauche](https://togithub.com/olivierdebauche) in [#1633](https://togithub.com/abetlen/llama-cpp-python/issues/1633) - fix(docker): Update Dockerfile build options from `LLAMA_` to `GGML_` by [@olivierdebauche](https://togithub.com/olivierdebauche) in [#1634](https://togithub.com/abetlen/llama-cpp-python/issues/1634) - feat: FreeBSD compatibility by [@yurivict](https://togithub.com/yurivict) in [#1635](https://togithub.com/abetlen/llama-cpp-python/issues/1635) ### [`v0.2.84`](https://togithub.com/abetlen/llama-cpp-python/blob/HEAD/CHANGELOG.md#0284) [Compare Source](https://togithub.com/abetlen/llama-cpp-python/compare/v0.2.83...v0.2.84) - feat: Update llama.cpp to [ggerganov/llama.cpp@`4730fac`](https://togithub.com/ggerganov/llama.cpp/commit/4730faca618ff9cee0780580145e3cbe86f24876) - fix: fix: Correcting run.sh filepath in Simple Docker implementation by [@mashuk999](https://togithub.com/mashuk999) in [#1626](https://togithub.com/abetlen/llama-cpp-python/issues/1626) ### [`v0.2.83`](https://togithub.com/abetlen/llama-cpp-python/blob/HEAD/CHANGELOG.md#0283) [Compare Source](https://togithub.com/abetlen/llama-cpp-python/compare/v0.2.82...v0.2.83) - feat: Update llama.cpp to [ggerganov/llama.cpp@`081fe43`](https://togithub.com/ggerganov/llama.cpp/commit/081fe431aa8fb6307145c4feb3eed4f48cab19f8) - feat: Add 'required' literal to ChatCompletionToolChoiceOption by [@mjschock](https://togithub.com/mjschock) in [#1597](https://togithub.com/abetlen/llama-cpp-python/issues/1597) - fix: Change repeat_penalty to 1.0 to match llama.cpp defaults by [@ddh0](https://togithub.com/ddh0) in [#1590](https://togithub.com/abetlen/llama-cpp-python/issues/1590) - fix(docs): Update README.md typo by [@ericcurtin](https://togithub.com/ericcurtin) in [#1589](https://togithub.com/abetlen/llama-cpp-python/issues/1589) - fix(server): Use split_mode from model settings by [@grider-withourai](https://togithub.com/grider-withourai) in [#1594](https://togithub.com/abetlen/llama-cpp-python/issues/1594) - feat(ci): Dockerfile update base images and post-install cleanup by [@Smartappli](https://togithub.com/Smartappli) in [#1530](https://togithub.com/abetlen/llama-cpp-python/issues/1530)huggingface/pytorch-image-models (timm)
### [`v1.0.8`](https://togithub.com/huggingface/pytorch-image-models/releases/tag/v1.0.8) [Compare Source](https://togithub.com/huggingface/pytorch-image-models/compare/v1.0.7...v1.0.8) ##### July 28, 2024 - Add `mobilenet_edgetpu_v2_m` weights w/ `ra4` mnv4-small based recipe. 80.1% top-1 @ 224 and 80.7 @ 256. - Release 1.0.8 ##### July 26, 2024 - More MobileNet-v4 weights, ImageNet-12k pretrain w/ fine-tunes, and anti-aliased ConvLarge models | model |top1 |top1\_err|top5 |top5\_err|param_count|img_size| |--------------------------------------------------------------------------------------------------|------|--------|------|--------|-----------|--------| | [mobilenetv4\_conv_aa_large.e230\_r448\_in12k_ft_in1k](http://hf.co/timm/mobilenetv4\_conv_aa_large.e230\_r448\_in12k_ft_in1k)|84.99 |15.01 |97.294|2.706 |32.59 |544 | | [mobilenetv4\_conv_aa_large.e230\_r384\_in12k_ft_in1k](http://hf.co/timm/mobilenetv4\_conv_aa_large.e230\_r384\_in12k_ft_in1k)|84.772|15.228 |97.344|2.656 |32.59 |480 | | [mobilenetv4\_conv_aa_large.e230\_r448\_in12k_ft_in1k](http://hf.co/timm/mobilenetv4\_conv_aa_large.e230\_r448\_in12k_ft_in1k)|84.64 |15.36 |97.114|2.886 |32.59 |448 | | [mobilenetv4\_conv_aa_large.e230\_r384\_in12k_ft_in1k](http://hf.co/timm/mobilenetv4\_conv_aa_large.e230\_r384\_in12k_ft_in1k)|84.314|15.686 |97.102|2.898 |32.59 |384 | | [mobilenetv4\_conv_aa_large.e600\_r384\_in1k](http://hf.co/timm/mobilenetv4\_conv_aa_large.e600\_r384\_in1k) |83.824|16.176 |96.734|3.266 |32.59 |480 | | [mobilenetv4\_conv_aa_large.e600\_r384\_in1k](http://hf.co/timm/mobilenetv4\_conv_aa_large.e600\_r384\_in1k) |83.244|16.756 |96.392|3.608 |32.59 |384 | | [mobilenetv4\_hybrid_medium.e200\_r256\_in12k_ft_in1k](http://hf.co/timm/mobilenetv4\_hybrid_medium.e200\_r256\_in12k_ft_in1k)|82.99 |17.01 |96.67 |3.33 |11.07 |320 | | [mobilenetv4\_hybrid_medium.e200\_r256\_in12k_ft_in1k](http://hf.co/timm/mobilenetv4\_hybrid_medium.e200\_r256\_in12k_ft_in1k)|82.364|17.636 |96.256|3.744 |11.07 |256 | - Impressive MobileNet-V1 and EfficientNet-B0 baseline challenges (https://huggingface.co/blog/rwightman/mobilenet-baselines) | model |top1 |top1\_err|top5 |top5\_err|param_count|img_size| |--------------------------------------------------------------------------------------------------|------|--------|------|--------|-----------|--------| | [efficientnet_b0.ra4\_e3600\_r224\_in1k](http://hf.co/timm/efficientnet_b0.ra4\_e3600\_r224\_in1k) |79.364|20.636 |94.754|5.246 |5.29 |256 | | [efficientnet_b0.ra4\_e3600\_r224\_in1k](http://hf.co/timm/efficientnet_b0.ra4\_e3600\_r224\_in1k) |78.584|21.416 |94.338|5.662 |5.29 |224 |\ | [mobilenetv1\_100h.ra4\_e3600\_r224\_in1k](http://hf.co/timm/mobilenetv1\_100h.ra4\_e3600\_r224\_in1k) |76.596|23.404 |93.272|6.728 |5.28 |256 | | [mobilenetv1\_100.ra4\_e3600\_r224\_in1k](http://hf.co/timm/mobilenetv1\_100.ra4\_e3600\_r224\_in1k) |76.094|23.906 |93.004|6.996 |4.23 |256 | | [mobilenetv1\_100h.ra4\_e3600\_r224\_in1k](http://hf.co/timm/mobilenetv1\_100h.ra4\_e3600\_r224\_in1k) |75.662|24.338 |92.504|7.496 |5.28 |224 | | [mobilenetv1\_100.ra4\_e3600\_r224\_in1k](http://hf.co/timm/mobilenetv1\_100.ra4\_e3600\_r224\_in1k) |75.382|24.618 |92.312|7.688 |4.23 |224 | - Prototype of `set_input_size()` added to vit and swin v1/v2 models to allow changing image size, patch size, window size after model creation. - Improved support in swin for different size handling, in addition to `set_input_size`, `always_partition` and `strict_img_size` args have been added to `__init__` to allow more flexible input size constraints - Fix out of order indices info for intermediate 'Getter' feature wrapper, check out or range indices for same. - Add several `tiny` < .5M param models for testing that are actually trained on ImageNet-1k |model |top1 |top1\_err|top5 |top5\_err|param_count|img_size|crop_pct| |----------------------------|------|--------|------|--------|-----------|--------|--------| |test_efficientnet.r160\_in1k |47.156|52.844 |71.726|28.274 |0.36 |192 |1.0 | |test_byobnet.r160\_in1k |46.698|53.302 |71.674|28.326 |0.46 |192 |1.0 | |test_efficientnet.r160\_in1k |46.426|53.574 |70.928|29.072 |0.36 |160 |0.875 | |test_byobnet.r160\_in1k |45.378|54.622 |70.572|29.428 |0.46 |160 |0.875 | |test_vit.r160\_in1k|42.0 |58.0 |68.664|31.336 |0.37 |192 |1.0 | |test_vit.r160\_in1k|40.822|59.178 |67.212|32.788 |0.37 |160 |0.875 | - Fix vit reg token init, thanks [Promisery](https://togithub.com/Promisery) - Other misc fixes ##### June 24, 2024 - 3 more MobileNetV4 hyrid weights with different MQA weight init scheme | model |top1 |top1\_err|top5 |top5\_err|param_count|img_size| |--------------------------------------------------------------------------------------------------|------|--------|------|--------|-----------|--------| | [mobilenetv4\_hybrid_large.ix_e600\_r384\_in1k](http://hf.co/timm/mobilenetv4\_hybrid_large.ix_e600\_r384\_in1k) |84.356|15.644 |96.892 |3.108 |37.76 |448 | | [mobilenetv4\_hybrid_large.ix_e600\_r384\_in1k](http://hf.co/timm/mobilenetv4\_hybrid_large.ix_e600\_r384\_in1k) |83.990|16.010 |96.702 |3.298 |37.76 |384 | | [mobilenetv4\_hybrid_medium.ix_e550\_r384\_in1k](http://hf.co/timm/mobilenetv4\_hybrid_medium.ix_e550\_r384\_in1k) |83.394|16.606 |96.760|3.240 |11.07 |448 | | [mobilenetv4\_hybrid_medium.ix_e550\_r384\_in1k](http://hf.co/timm/mobilenetv4\_hybrid_medium.ix_e550\_r384\_in1k) |82.968|17.032 |96.474|3.526 |11.07 |384 | | [mobilenetv4\_hybrid_medium.ix_e550\_r256\_in1k](http://hf.co/timm/mobilenetv4\_hybrid_medium.ix_e550\_r256\_in1k) |82.492|17.508 |96.278|3.722 |11.07 |320 | | [mobilenetv4\_hybrid_medium.ix_e550\_r256\_in1k](http://hf.co/timm/mobilenetv4\_hybrid_medium.ix_e550\_r256\_in1k) |81.446|18.554 |95.704|4.296 |11.07 |256 | - florence2 weight loading in DaViT modeltqdm/tqdm (tqdm)
### [`v4.66.5`](https://togithub.com/tqdm/tqdm/releases/tag/v4.66.5): tqdm v4.66.5 stable [Compare Source](https://togithub.com/tqdm/tqdm/compare/v4.66.4...v4.66.5) - support `ncols` auto-detection on FreeBSD ([#1602](https://togithub.com/tqdm/tqdm/issues/1602) <- [https://github.com/casperdcl/git-fame/issues/98](https://togithub.com/casperdcl/git-fame/issues/98)) - fix Python 3.13 CLI ([#1594](https://togithub.com/tqdm/tqdm/issues/1594) <- [#1585](https://togithub.com/tqdm/tqdm/issues/1585)) - fix Python 3.13 tests ([#1595](https://togithub.com/tqdm/tqdm/issues/1595) <- [https://github.com/python/cpython/issues/117536#issuecomment-2036883124](https://togithub.com/python/cpython/issues/117536#issuecomment-2036883124)) - misc framework updates ([#1602](https://togithub.com/tqdm/tqdm/issues/1602)) - add official Python 3.12 support - bump deps ([https://github.com/NiklasRosenstein/pydoc-markdown/issues/329](https://togithub.com/NiklasRosenstein/pydoc-markdown/issues/329), [https://github.com/tikitu/jsmin/pull/44](https://togithub.com/tikitu/jsmin/pull/44))encode/uvicorn (uvicorn)
### [`v0.30.5`](https://togithub.com/encode/uvicorn/blob/HEAD/CHANGELOG.md#0305-2024-08-02) [Compare Source](https://togithub.com/encode/uvicorn/compare/0.30.4...0.30.5) ##### Fixed - Don't close connection before receiving body on H11 ([#2408](https://togithub.com/encode/uvicorn/issues/2408)) ### [`v0.30.4`](https://togithub.com/encode/uvicorn/blob/HEAD/CHANGELOG.md#0304-2024-07-31) [Compare Source](https://togithub.com/encode/uvicorn/compare/0.30.3...0.30.4) ##### Fixed - Close connection when `h11` sets client state to `MUST_CLOSE` ([#2375](https://togithub.com/encode/uvicorn/issues/2375))Configuration
📅 Schedule: Branch creation - "before 4am on Monday" (UTC), Automerge - At any time (no schedule defined).
🚦 Automerge: Enabled.
♻ Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.
👻 Immortal: This PR will be recreated if closed unmerged. Get config help if that's undesired.
This PR has been generated by Renovate Bot.