This page contains open issues tagged with the label good first issue. It gathers issues from some famous libraries(at least for me). Let me knowif there are others you would like to see here 🍕
update_issues.py can be modified to return the data in json or to csv.
To run the script locally:
- create an
.envfile in the root directory, and add
ACCESS_TOKEN=<YOUR_ACCESS_TOKEN>
USERNAMES=<REPO_OWNER1>,<REPO_OWNER2>,<REPO_OWNER3>,...
- How to create and manage GitHub API access tokens in this article.
If you are wondering where to begin in the journey contributing to open-source projects, these are good articles to get you started! ⛵
| Repo | Title | Comments |
|---|
| pytorch/csprng | Clone torchvision .circleci/regenerate.py framework to automate config.yml generation | 0 |
| pytorch/csprng | Add clang-format to CircleCI config | 0 |
| pytorch/csprng | Setup CMake build | 0 |
| pytorch/csprng | Improve test_geometric | 0 |
| pytorch/csprng | Delete OffsetCalculator.cuh and THCIntegerDivider.cuh and replace them with PyTorch headers | 1 |
| pytorch/csprng | ZeroDivisionError: float division by zero in test_cpu_parallel | 3 |
| Repo | Title | Comments |
|---|
| pytorch/glow | [TensorLayout] Add a way to specify canonical layout propagation to NodeGen | 0 |
| pytorch/glow | [Runtime] Expand Yaml DeviceConfig loading to include HostConfig | 0 |
| pytorch/glow | [runtime] add device memory size configuration to the DeviceConfig::parameters map on all backends | 0 |
| pytorch/glow | resnet-* example code should reuse Tensor memory for images | 0 |
| pytorch/glow | Example binaries should handle HostManager active request limit | 0 |
| pytorch/glow | [ONNXIFI] Add additional precondition checks to model loaders | 0 |
| pytorch/glow | [ONNXIFI][GraphAPI] Add "orErr" methods for graph creation api | 0 |
| pytorch/glow | ConvBench should use the layers of ResNet | 0 |
| pytorch/glow | ConvBench should use libjit_convolution_D8KKC8 | 0 |
| pytorch/glow | [CPU] Remove the need for Rescale for the CPUMaxSplat | 1 |
| pytorch/multipy | Create a Contributing.md | 1 |
| pytorch/multipy | Add Poetry build to multipy | 1 |
| pytorch/glow | New files added under glow/test/models/ don't get copied to build folder until clean build is performed. | 2 |
| pytorch/glow | Make all useful data structures in Glow easy to dump in a textual form | 2 |
| pytorch/glow | [Importers] Move "run and compare" tests from importer tests to operator tests | 2 |
| pytorch/glow | Need to canonicalize or optimize high-dim concat to concat+transpose | 2 |
| tensorflow/minigo | Support more GTP commands | 3 |
| pytorch/glow | Move layout information from nodes constructor' to their Glow types | 3 |
| pytorch/glow | [habana] Implement backend specific verifications | 3 |
| pytorch/glow | [GraphOptz] Sink Quantize below Reshape | 4 |
| pytorch/glow | Fuse RescaleQuantized in the ChannelwiseQuantizedConv | 5 |
| pytorch/glow | Bring back the sinking RELU below Pool optimization | 5 |
| pytorch/glow | Use gtest-parallel | 5 |
| godotengine/godot | _OnTreeItemSelected() signal receiver method using C# doesn't allow adding of items | 5 |
| pytorch/glow | Glow should provide its own implementation of a random number generator to ensure deterministic testing and avoid platform-dependent differences | 6 |
| pytorch/glow | Make the execution of the Glow compiler more deterministic between multiple runs | 9 |
| opencv/opencv_contrib | reg: broken python sample | 9 |
| godotengine/godot | [Godot v4.4] Gizmos exponentially increasing Draw Calls and Objects when turned on | 11 |
| opencv/opencv_contrib | selectROI pointer almost invisible on certain colours | 14 |
| pytorch/glow | [ONNXModelLoader] Support for LogSoftmax operator is missing | 16 |
| pytorch/glow | Use explicit std::string conversions instead of implicit llvm::StringRef->std::string conversions to prepare Glow for builds using LLVM >=11 | 18 |
| godotengine/godot | Using the RenderingServer to add a texture to the scene, the texture appears at bad position during one frame | 24 |
| godotengine/godot | You're breathtaking! | 30 |
| godotengine/godot | PointLight2D doesn't respect Nearest texture filter for normal-maps or shadows | 30 |
| opencv/opencv | HDR tutorial result is different from the code running. | 32 |
| tensorflow/tensorflow | Feature Request: GPUOptions for Go binding | 109 |
| godotengine/godot | [TRACKER] Unit tests to add or improve | 227 |
| Repo | Title | Comments |
|---|
| ansible/pylibssh | [TODO] Ensure release builds don't include debug symbols | 0 |
| Repo | Title | Comments |
|---|
| godotengine/emacs-gdscript-mode | Rewrite gdscript-imenu to provide GDScript-specific tables | 0 |
| godotengine/emacs-gdscript-mode | Buffer does not revert or update instantly after formatting buffer | 3 |
| godotengine/emacs-gdscript-mode | Address compile warnings | 6 |
| Repo | Title | Comments |
|---|
| godotengine/godot-demo-projects | Update plugins demo to use custom Resources | 0 |
| godotengine/godot-benchmarks | [Call to Action] Benchmarks for the benchmark server | 5 |
| godotengine/godot-benchmarks | [TRACKER] Benchmarks to create | 11 |
| Repo | Title | Comments |
|---|
| ansible/receptor | append conntype string representation in status output | 0 |
| ansible/receptor | add work status command to receptorctl | 1 |
| ansible/receptor | quality of life -- order work list output based on most recent work units | 2 |
| Repo | Title | Comments |
|---|
| monero-project/monero-site | fragment linking to specific sections | 1 |
| django/birthday20 | Vendor third-party assets (leaflet) | 4 |
| Repo | Title | Comments |
|---|
| pytorch/serve | The captum for bert notebook needs update | 0 |
| pytorch/serve | Enable naked DIR test case for windows environment | 0 |
| pytorch/serve | [Educational] Typing .py files iteratively with mypy | 1 |
| pytorch/serve | Built-in handler for regression | 1 |
| pytorch/serve | permission error while trying to remove previous artifact files in /tmp/ | 3 |
| pytorch/serve | Torchserve M1 support detailed plan | 4 |
| pytorch/serve | add support for SeTFiT models | 4 |
| pytorch/serve | [BUG] Dockerfile with nvidia/cuda:10.2-cudnn7-runtime-ubuntu18.04 gives ascii codec error | 4 |
| pytorch/serve | how does default_response_timeout work? | 6 |
| pytorch/serve | [Suggestion] Loading torch-model-archiver arguments from YAML | 6 |
| pytorch/serve | Handle invalid input intelligently ！ | 7 |
| Repo | Title | Comments |
|---|
| huggingface/transformers.js | Is 'aggregation_strategy' parameter available for token classification pipeline? | 2 |
| huggingface/transformers.js | [Feature request] Return offset mapping using tokenizer | 2 |
| pytorch/ci-hud | Restore "Show service jobs" checkbox to GitHubStatusDisplay.js | 2 |
| huggingface/transformers.js | [Doc request] Add an example guide of how to use it in Svelte (and deploy to HF Spaces) | 7 |
| huggingface/transformers.js | [Feature request] Logging Level and Progress Bar for Model Downloads | 9 |
| Repo | Title | Comments |
|---|
| ansible/galaxy_collection | Replace order of roles | 0 |
| ansible/galaxy_collection | infra.ah_configuration.publish leaves clone dirs behind | 0 |
| ansible/awx-operator | Add multi-arch build target | 1 |
| ansible/product-demos | Add teardown template/workflow for "Deploy Cloud Stack in AWS" workflow | 1 |
| ansible/awx-operator | annotations not applied to postgres statefulset | 2 |
| ansible/awx-operator | Inflated Total Host Count on Dashboard when Using Constructed Inventory with Existing Inventories | 2 |
| ansible/awx-operator | Extend README with external logging examples | 2 |
| ansible/receptor-collection | Support for Debian 12 | 2 |
| ansible/awx-operator | Inability to overwrite namespace when using awx-operator chart | 3 |
| ansible/awx-operator | Password set via UI gets overwritten by initialize_django playbook | 5 |
| Repo | Title | Comments |
|---|
| tensorflow/probability | providing sample weights to glm.fit() and glm.fit.sparse() | 0 |
| tensorflow/probability | Sigmoid belief network example/tutorial | 0 |
| tensorflow/probability | Logistic factorial analysis example/tutorial | 0 |
| tensorflow/swift-models | Add documentation to the Gym models to specify additional dependencies for DQN | 0 |
| tensorflow/swift-models | [BERT-CoLA]: Use common file extraction functions | 0 |
| tensorflow/swift-models | Verify BERT variants | 0 |
| pytorch/botorch | [Feature Request] Allow negative indices to fixed_features in optimizers | 0 |
| pytorch/opacus | Update grad sample algorithm for Convolution | 0 |
| tensorflow/probability | Feature Request: Add crps method to tfp.distributions | 1 |
| tensorflow/probability | Keyword argument overload (?) in tfb.real_nvp_default_template | 1 |
| tensorflow/probability | Item response theory model example/tutorial | 1 |
| tensorflow/probability | Bayesian neural network tutorial | 1 |
| tensorflow/swift-models | Add logging callbacks for training statistics | 1 |
| tensorflow/probability | Support for SeparableConv1DFlipout | 2 |
| tensorflow/probability | NotImplementedError: mean is not implemented: Categorical | 2 |
| tensorflow/probability | Always-on dropout layer | 2 |
| tensorflow/probability | DenseFlipOut layer | 2 |
| tensorflow/probability | Categorical autoregressive distribution with LSTM example/tutorial | 2 |
| tensorflow/swift-models | Better error message on unzip failure | 2 |
| tensorflow/swift-models | Find a more descriptive word for numericalized | 2 |
| tensorflow/swift-models | Rewrite long nested Sequential types using Sequential{N} typealiases | 2 |
| tensorflow/swift-models | [BERT-CoLA] Data source resolution | 2 |
| tensorflow/swift-models | Object detection examples | 2 |
| pytorch/functorch | Add vmap support for PyTorch operators | 2 |
| pytorch/opacus | Double-charging privacy in DC-GAN example | 2 |
| tensorflow/adanet | Allow one to forward features to predictions | 3 |
| tensorflow/probability | Feature Request: Alternative Horseshoe Parameterization as tfp distribution | 3 |
| tensorflow/probability | Please add example on NUTS sampling from posterior for Bayesian Logistic Regression | 3 |
| tensorflow/probability | Feature request: Tweedie distribution | 3 |
| tensorflow/probability | CholeskyWishart distribution | 3 |
| tensorflow/probability | improve text and visuals in Bayesian Gaussian mixture model tutorial | 3 |
| tensorflow/swift | Installation verification tool | 3 |
| tensorflow/probability | A question about target_log_prob_fn | 4 |
| tensorflow/probability | Provide a resimulation Metropolis TransitionKernel | 4 |
| tensorflow/swift-models | Translation example | 4 |
| pytorch/botorch | [Feature Request] InputDataWarnings are are confusing | 4 |
| tensorflow/probability | Laplace approximation | 5 |
| tensorflow/probability | Missing Inverse Wishart - redefine Wishart? | 5 |
| tensorflow/probability | Categorical features in logistic regression | 5 |
| tensorflow/probability | Stochastic block model example/tutorial | 5 |
| tensorflow/probability | BFGS user should be able to pass parameters to the line search | 6 |
| tensorflow/probability | Tutorial on Variational Inference | 6 |
| tensorflow/swift-models | Speech to text example | 7 |
| tensorflow/probability | Any plans to update the examples to TensorFlow 2? | 8 |
| tensorflow/probability | Sample different weights for examples within a batch | 9 |
| huggingface/huggingface-gemma-recipes | 📣 Call for contributions! | 10 |
| tensorflow/probability | [Feature Request] tensorflow_probability.distributions.GeneralizedGamma | 11 |
| tensorflow/probability | Feature Request: Efficient Poisson Binomial PMF/CDF in tfp | 13 |
| tensorflow/probability | [Feature Request] Zero-Inflated Poisson and Negative Binomial distributions | 13 |
| tensorflow/swift-models | UNet Example | 15 |
| huggingface/huggingface-llama-recipes | Call for contributions | 20 |
| Repo | Title | Comments |
|---|
| huggingface/agents-course | Improve diagrams in unit 2.1 | 4 |
| Repo | Title | Comments |
|---|
| zeromicro/goctl-swagger | support for openapi 3.0 | 0 |
| godotengine/godot-asset-library | Categories need a description (tooltip?) when hovered/selected | 1 |
| godotengine/godot-asset-library | keep the content of the submition form when its not validated, or try some ajax realtime pre-validation | 1 |
| godotengine/godot-asset-library | A way to cancel edit requests | 3 |
| godotengine/godot-asset-library | Sanitize inputs for Asset data/fields, like URLs | 5 |
| Repo | Title | Comments |
|---|
| django/djangoproject.com | Github login doesn't work for community pages | 3 |
| django/djangoproject.com | Documentation table of contents is hard to reach on mobile devices | 4 |
| django/djangoproject.com | Add link to 'Stable' doc version in the Warning Alert on top | 5 |
| django/djangoproject.com | fundraising: 'customer.subscription.deleted' webhook event always gets 404 response | 5 |
| django/djangoproject.com | Add link to new minutes repo | 6 |
| django/djangoproject.com | Docs for old supported versions should indicate that a newer version is available | 6 |
| django/djangoproject.com | Support switching languages on non-docs sites | 6 |
| django/djangoproject.com | Improvements to the Corporate Sponsor Experience | 7 |
| django/djangoproject.com | Improve Documentation by having list of topics fixed on the left and table of contents on the right | 7 |
| django/djangoproject.com | Use noindex meta tag or header, not robots.txt, to block untranslated docs pages | 7 |
| django/djangoproject.com | https://dashboard.djangoproject.com/foundation/corporate-membership/ > RETURNS 404 | 8 |
| django/djangoproject.com | Remove non-canonical docs versions from sitemap.xml | 8 |
| django/djangoproject.com | Make right sidebar independent from main content | 12 |
| django/djangoproject.com | New content: Accessibility statement | 13 |
| django/djangoproject.com | Improve 404 page | 13 |
| django/djangoproject.com | Localize the fundraising app | 17 |
| django/djangoproject.com | Add Click to copy text for commands present in the official Django website. | 19 |
| Repo | Title | Comments |
|---|
| huggingface/lerobot | Ensure the configuration module passes MyPy type checks | 0 |
| huggingface/lerobot | Ensure the optimization module passes MyPy type checks | 0 |
| huggingface/lerobot | Ensure the environments module passes MyPy type checks | 0 |
| huggingface/lerobot | Ensure the utilities module passes MyPy type checks | 0 |
| huggingface/lerobot | Ensure the teleoperators module passes MyPy type checks | 0 |
| huggingface/lerobot | Ensure the processing module passes MyPy type checks | 0 |
| huggingface/lerobot | Ensure the cameras module passes MyPy type checks | 0 |
| huggingface/lerobot | Ensure the motors module passes MyPy type checks | 0 |
| huggingface/lerobot | Ensure the datasets module passes MyPy type checks | 0 |
| huggingface/lighteval | [FT] showing count in Markdown summary table | 0 |
| huggingface/lighteval | [FT] add py.typed so lighteval can work with type checkers | 0 |
| huggingface/lighteval | [FT] Manage script and language in the Language enum | 0 |
| huggingface/lighteval | [BUG] Optimize tokenization | 0 |
| huggingface/lighteval | [FT] Add tests for VLLMModel base methods | 0 |
| huggingface/lighteval | [FT] Build in a way to specify specific IDs/Lines in Dataset to use as few-shot examples in the same split | 0 |
| huggingface/nanotron | [Feature] Use CUDA event for measuring elasped time | 0 |
| huggingface/nanotron | [Feature] Refactor ParallelContext.world_rank_matrix | 0 |
| huggingface/optimum-executorch | Add benchmarking numbers for more models | 0 |
| ansible/ansible-dev-tools | Ensure tests are passing with devspaces container | 0 |
| ansible/ansible-jupyter-kernel | Add support for handlers | 0 |
| ansible/ansible-jupyter-kernel | Add a #library cell type | 0 |
| ansible/awx | [ui_next] Add All/Any workflow viz indication to Workflow Viz Legend/Key | 0 |
| ansible/dispatcherd | Enforce UUID format, add "origin" information in message data | 0 |
| ansible/dispatcherd | Do some minor linting of test files | 0 |
| ansible/django-ansible-base | ansible_base.lib.utils.models.diff takes unused sanitize_encrypted param and has outdated return docstring | 0 |
| tensorflow/moonlight | Follow the MusicXML schema | 0 |
| tensorflow/moonlight | Export stem directions in MusicXML | 0 |
| tensorflow/moonlight | Chord duration correction | 0 |
| tensorflow/moonlight | Export the key signature to MusicXML | 0 |
| pytorch/ao | [fp8 blockwise training] try using torch._scaled_mm instead of Triton kernels for fp8 gemms | 0 |
| pytorch/ao | TorchAO ROCM tests are taking a long time | 0 |
| pytorch/ao | Add _apply_fn_to_data in AOBaseClass | 0 |
| pytorch/ao | Remove docstring duplications | 0 |
| pytorch/examples | imagenet benchmark measure elapsed time function issue | 0 |
| pytorch/examples | Why there is no tanh activation at the end of TransformerNet? | 0 |
| pytorch/examples | Distributed doesn't work with env:// init_method | 0 |
| pytorch/examples | Is the loss of the first word covered during the language model evaluation? | 0 |
| pytorch/examples | The casting problem on the result of the model. | 0 |
| pytorch/examples | Example code (mnist) does not work under torch.utils.bottleneck | 0 |
| pytorch/executorch | Support getting the delegated payload | 0 |
| pytorch/executorch | [Android] Rename JNI cpp file | 0 |
| pytorch/executorch | [CMake] Potentially duplicated srcs in llama_runner build | 0 |
| pytorch/executorch | [Android] Add method metadata report API in Module.java | 0 |
| pytorch/PiPPy | Support Segformer models in HF tests | 0 |
| pytorch/PiPPy | Support ConvNext models in HF tests | 0 |
| pytorch/PiPPy | Support Wav2Vec2 models in HF tests | 0 |
| pytorch/PiPPy | Support ResNet models in HF tests | 0 |
| pytorch/PiPPy | [spmd] DistributedTensor doesn't implement is_pinned | 0 |
| pytorch/PiPPy | Support Donut SWIN models in HF tests | 0 |
| pytorch/PiPPy | Create mypy.ini | 0 |
| pytorch/PiPPy | Support DistilBert models in HF tests | 0 |
| pytorch/PiPPy | Support Deberta models in HF tests | 0 |
| pytorch/PiPPy | Support HubertForSequenceClassification models in HF tests | 0 |
| pytorch/PiPPy | Support LxmertForPreTraining models in HF tests | 0 |
| pytorch/PiPPy | Support Speech2Text models in HF tests | 0 |
| pytorch/PiPPy | Support CLIP models in HF tests | 0 |
| pytorch/PiPPy | Support SWIN models in HF tests | 0 |
| pytorch/rl | [DO NOT CLOSE] Call for contributions | 0 |
| pytorch/tensordict | [Feature Request] Deprecate _run_checks in favour of TensorDict._checked | 0 |
| pytorch/torcharrow | Making BaseColumn::genericUnaryUDF and the family free functions | 0 |
| pytorch/torcharrow | Call type inferring in interop.from_pylist | 0 |
| pytorch/torcharrow | Move torcharrow/test to test | 0 |
| pytorch/torchcodec | More robust metadata serialization | 0 |
| pytorch/torchtune | Log config to output only on rank zero | 0 |
| pytorch/torchtune | Add add_end_token to the Qwen Models | 0 |
| pytorch/torchtune | Add add_end_token to the Gemma Tokenizer | 0 |
| pytorch/xla | [torch_xla] scan_layers fails if the layer have no weights | 0 |
| pytorch/xla | Op info test for bmm .. ceil | 0 |
| pytorch/xla | Op info test for rsub .. searchsorted | 0 |
| pytorch/xla | Op info test for norm .. pca_lowrank | 0 |
| pytorch/xla | Op info test for nn.functional.softmin .. nonzero_static | 0 |
| pytorch/xla | Op info test for nn.functional.group_norm .. nn.functional.layer_norm | 0 |
| pytorch/xla | Op info test for nn.functional.conv_transpose3d .. nn.functional.ctc_loss | 0 |
| pytorch/xla | Op info test for linalg.ldl_factor_ex | 0 |
| pytorch/xla | Op info test for linalg.householder_product | 0 |
| pytorch/xla | Op info test for linalg.cholesky_ex | 0 |
| pytorch/xla | Op info test for linalg.cholesky | 0 |
| huggingface/dataset-viewer | Use revision_exists (hfh) | 1 |
| huggingface/diffusers | Do we have any script covert from hf format to orginal format? | 1 |
| huggingface/lerobot | Ensure the model module passes MyPy type checks | 1 |
| huggingface/lerobot | Ensure the transport module passes MyPy type checks | 1 |
| huggingface/lighteval | [FT] Add tests for nanotron | 1 |
| huggingface/lighteval | [BUG] Misleading metric type of LightevalTaskConfig | 1 |
| huggingface/lighteval | [EVAL] Big-Bench Extra Hard (BBEH) | 1 |
| huggingface/lighteval | [FT] LiteLLM concurrency parameters hard-coded | 1 |
| huggingface/nanotron | Add Debug utility to be able to preview first samples used for training | 1 |
| huggingface/nanotron | We don't save checkpoint after training ends | 1 |
| huggingface/nanotron | FEAT: Support 1.58-bit LLMs training | 1 |
| huggingface/nanotron | [Feature Request] Add simple communications benchmarks to the repo | 1 |
| ansible/ansible-lint | Linter should consider the same paths that ansible.builtin.import_tasks uses to import relative paths | 1 |
| ansible/ansible-navigator | Many params missing change after initial = False | 1 |
| ansible/awx | Scheduled Job prompted extra_vars overwritten by JT extra_vars | 1 |
| ansible/awx | Inaccurate error handling on Input Group Integer fields (i.e. Org Max Hosts field) | 1 |
| ansible/django-ansible-base | Several unordered querysets | 1 |
| tensorflow/agents | Tutorial 1: Train a Deep Q Network with TF-Agents and the use of a driver | 1 |
| tensorflow/quantum | Documentation Format | 1 |
| pytorch/ao | Refactor torchao and tests to use model architectures from torchao.testing.model_architectures | 1 |
| pytorch/ao | Add weight tensor-wise scaling for INT8 quantized and mixed-precision training | 1 |
| pytorch/ao | MoE example | 1 |
| pytorch/ao | Self compressing neural networks | 1 |
| pytorch/data | Modify IterKeyZipper to accept any number (2+) of IterDataPipe | 1 |
| pytorch/examples | Update all examples to latest versions of pytorch | 1 |
| pytorch/examples | Video classification example | 1 |
| pytorch/examples | Resume will override learning rate on command line. | 1 |
| pytorch/examples | Add nn.SyncBatchNorm in ImageNet example | 1 |
| pytorch/examples | mnist example contains bad practices leading to wrongful results | 1 |
| pytorch/examples | A question about weight initialization of embedding layer. | 1 |
| pytorch/examples | Changes needed to run DCGAN 32x32 | 1 |
| pytorch/examples | ImageFolder doc should clarify 1. order that images returned in 2. that all classes are concatenated into a single list | 1 |
| pytorch/examples | Doc comment on accuracy method in imagenet example, incorrect? | 1 |
| pytorch/executorch | Add pybindings for text LLM runner | 1 |
| pytorch/executorch | Support dim order in Java Tensor | 1 |
| pytorch/executorch | Add dynamism tests for xnnpack model tests | 1 |
| pytorch/executorch | Add test to ensure scalar_type.fbs remains in sync across the executorch repo | 1 |
| pytorch/executorch | Implement load_into for the shared_ptr data loader | 1 |
| pytorch/executorch | Remove the outdated warning in building from source documentation | 1 |
| pytorch/executorch | Test exported models in Python | 1 |
| pytorch/executorch | Add timestamps for pte generation in CI | 1 |
| pytorch/executorch | Format CMakeLists.txt | 1 |
| pytorch/executorch | [Android] instrumentation test use models from storage, not bundled with apk | 1 |
| pytorch/executorch | [Android] Use generic JNI instead of fbjni | 1 |
| pytorch/helion | Limit block sizes in autotuning because of triton max numel range | 1 |
| pytorch/PiPPy | Support LayoutLM models in HF tests | 1 |
| pytorch/pytorch | [inductor] nan_asserts doesn't work for FP8, "RuntimeError: "isinf" not implemented for 'Float8_e4m3fn'" | 1 |
| pytorch/text | torchtext main branch doesn't support pytorch2.0 | 1 |
| pytorch/torcharrow | Default aggregation functions delegates to Arrow Compute | 1 |
| pytorch/torchchat | GeneratorArgs.is_torchtune_model is a misnomer | 1 |
| pytorch/torchcodec | Add memory consumption decoder benchmarks | 1 |
| pytorch/torchtitan | Ability to train based on epoch | 1 |
| pytorch/torchtune | Update ReadMe Optimization Flags | 1 |
| pytorch/xla | xs.make_mesh | 1 |
| pytorch/xla | torch_xla scan forces inputs to be differentiable | 1 |
| pytorch/xla | [hard] Op info test for special.airy_ai | 1 |
| pytorch/xla | Op info test for triangular_solve .. unfold | 1 |
| pytorch/xla | Op info test for nn.functional.pixel_shuffle .. nn.functional.scaled_dot_product_attention | 1 |
| pytorch/xla | Op info test for linalg.matrix_norm | 1 |
| pytorch/xla | Op info test for linalg.lu_factor_ex | 1 |
| pytorch/xla | Op info test for linalg.lu_factor | 1 |
| pytorch/xla | Op info test for index_reduce | 1 |
| pytorch/xla | [hard] Op info test for histogramdd | 1 |
| huggingface/dataset-viewer | use the ROW_IDX_COLUMN constant name instead of copying the value everywhere | 2 |
| huggingface/dataset-viewer | Use "Sign-In with HF" instead of token in admin-UI | 2 |
| huggingface/lerobot | Ensure the robots module passes MyPy type checks | 2 |
| huggingface/lighteval | Call for contributions: Translate lighteval's doc into Chinese | 2 |
| huggingface/nanotron | [Feature] Asyncronous Serialization | 2 |
| huggingface/nanotron | [Feature Request] Support Data Streaming for faster training of large models | 2 |
| huggingface/optimum | Add all available ONNX models to ORTConfigManager | 2 |
| huggingface/transformers | [Performance] Tracking open Issues and PRs (pytorch transformers) | 2 |
| ansible/ansible-documentation | Link "see also" sections as recommended in the forum | 2 |
| ansible/ansible-navigator | Switch to the use of pathlib where possible | 2 |
| ansible/awx | cancelled vs canceled - standardize spelling in codebase | 2 |
| ansible/awx | Empty value in execution_environment should remove selections (ansible.controller.job_template) | 2 |
| ansible/awx | Issue label for incompliance with Ansible | 2 |
| ansible/awx | New credentials type to provide VAULT_ADDR and VAULT_TOKEN env vars | 2 |
| ansible/awx | Expose the value of "Execution Node" as a variable when running a job from Ansible Tower. | 2 |
| ansible/awx | "Playbook not found for project" error when the playbook exist but is invalid | 2 |
| ansible/awx | Jobs canceled before starting should not show a "finished" time | 2 |
| ansible/awx | Make Info Metrics more easy to query | 2 |
| ansible/awx | weekly dashboard view does not have sufficient granularity | 2 |
| tensorflow/graphics | Tangential camera distortion/undistortion | 2 |
| pytorch/ao | Deprecate and remove subclass.py, dynamic_quant.py, and weight_only.py | 2 |
| pytorch/ao | Unittests Migration Progress | 2 |
| pytorch/ao | [QAT] Low-bit FSDP all-gather for QAT | 2 |
| pytorch/ao | [MX | Triton] Create MX matmul op using new scaled_dot op in Triton | 2 |
| pytorch/ao | Support for GrokAdamW | 2 |
| pytorch/ao | Tensor subclass boilerplate can be consolidated | 2 |
| pytorch/ao | [RFC] Intx Tensor Subclasses Quantization | 2 |
| pytorch/examples | word_language_model with torch.nn.modules.transformer | 2 |
| pytorch/examples | [Warning: an output with one or more elements was resized since it had shape[xxxx],]](pytorch/examples#937) | 2 |
| pytorch/examples | Input type and weight type should be the same | 2 |
| pytorch/examples | please use longer names for variables than 2-3 letter acronyms | 2 |
| pytorch/examples | VAE reconstruction loss (BCE) | 2 |
| pytorch/examples | Imagenet training extremely low gpu utilization | 2 |
| pytorch/executorch | [Arm] Size-test: Run the binary on CI | 2 |
| pytorch/executorch | Consolidation of Selective Build APIs for OSS | 2 |
| pytorch/executorch | Support Standalone Batch Norm | 2 |
| pytorch/executorch | Re-inplace slice_copy with slice | 2 |
| pytorch/executorch | Add torchao kernels to xcframework | 2 |
| pytorch/executorch | Need a feature to get etdump while running LLAMA model on qnn with qnn_llama_runner | 2 |
| pytorch/executorch | [Request impl] Gracefully error out in ETDump | 2 |
| pytorch/PiPPy | Make RemoteInterpreter use the full implementation of Interpreter.run | 2 |
| pytorch/pytorch | test/distributed/_tools/test_sac_ilp.py::TestSACILP::test_sac_ilp_case1 is seemingly not run in CI | 2 |
| pytorch/rl | [Feature Request] Make sure that all losses work with tensorclasses and regular tensors | 2 |
| pytorch/rl | [Feature Request] TensorSpec is_in methods should check the dtype of val | 2 |
| pytorch/text | Convert iterator-style raw datasets to map-style raw datasets | 2 |
| pytorch/torchchat | torchtune as an optional dependency: Lazy Import | 2 |
| pytorch/torchchat | Slimming down torchchat: Replace replace_attention_with_custom_sdpa_attention() with ET's implementation | 2 |
| pytorch/torchcodec | Create unit tests based on variable FPS video | 2 |
| pytorch/torchtune | [Multi-Node] Test recipes/knowledge_distillation_distributed.py on a multi-node setup | 2 |
| pytorch/xla | Op info test for special.scaled_modified_bessel_k0 .. squeeze | 2 |
| pytorch/xla | Op info test for nn.functional.feature_alpha_dropout .. nn.functional.grid_sample | 2 |
| pytorch/xla | Op info test for linalg.matrix_power | 2 |
| pytorch/xla | Op info test for linalg.ldl_solve | 2 |
| huggingface/accelerate | [Community Contributions] examples on distributed inference using 🤗 Accelerate | 3 |
| huggingface/dataset-viewer | Use CONSTANT_LIST.copy in list config fieds | 3 |
| huggingface/huggingface_hub | Add options to the "delete-cache" command | 3 |
| huggingface/lerobot | Ensure the policy module passes MyPy type checks | 3 |
| huggingface/optimum-habana | Add support for max_length in run_generation | 3 |
| huggingface/transfer-learning-conv-ai | RuntimeError: shape '[-1, 2, 34]' is invalid for input of size 61710 | 3 |
| pandas-dev/pandas | DOC: Clarify parentheses vs. brackets | 3 |
| ansible/ansible-documentation | Incorporate 'shared snippets' for the collection docs into the main rst files | 3 |
| ansible/awx | Add option to force the user to reset its password at first login | 3 |
| ansible/awx | Fix Delinea imports and import test | 3 |
| ansible/awx | Cleanup Job Details. Configure days retention for the particular job entities. | 3 |
| ansible/awx | Improve user experience of copy pasting FQDNs from jobs | 3 |
| ansible/awx | Non descriptive error when podman is missing | 3 |
| ansible/awx | AWX Credential Create: ssh_key_data fails to detect/sanitize newlines in keys and 500's with no useable output | 3 |
| ansible/awx | Add "nodes" to "Relaunch on" when relaunching jobs after failure - users think it relates to failed tasks and not nodes | 3 |
| ansible/awx | Worker failed to run task awx.main.tasks.system.purge_old_stdout_files | 3 |
| ansible/awx | System Auditors can check list items | 3 |
| ansible/awx | Show more host information in host filter lookup list | 3 |
| ansible/awx | Inconsistent capitalization across ui-next | 3 |
| tensorflow/addons | Benchmark test for rrelu fails on windows | 3 |
| tensorflow/graphics | Add skew to projective camera | 3 |
| tensorflow/model-card-toolkit | Prefer single quote strings model-card-toolkit | 3 |
| tensorflow/similarity | Add a ranking example | 3 |
| pytorch/ao | [not user facing] Split implements and implements_torch_function | 3 |
| pytorch/ao | [sparse] Add GPTQ support for sparse-marlin | 3 |
| pytorch/ao | Tensor Parallelism Support for AffineQuantizedTensor | 3 |
| pytorch/ao | [llama] Use horizontal fusion trick from Attention for FeedForward | 3 |
| pytorch/data | scan support | 3 |
| pytorch/examples | OSError: [Errno 22] Invalid argument: | 3 |
| pytorch/examples | MNIST test_loader not being used as intended (wrong)! | 3 |
| pytorch/examples | Actor critic example not using discount rate properly | 3 |
| pytorch/examples | "omit freeze_support" | 3 |
| pytorch/executorch | torch.split fails in to_edge | 3 |
| pytorch/executorch | Get rid of fixed number cmake --build -j in sh and docs | 3 |
| pytorch/executorch | Consolidate executor_runners | 3 |
| pytorch/ignite | Try to replace conda/setup-python with astral-uv if faster | 3 |
| pytorch/pytorch | Remove tensor variable default method fallthrough | 3 |
| pytorch/pytorch | [export] Schematize nn_module_stack serialization | 3 |
| pytorch/torchchat | Improve Tokenizer New Type Onboarding | 3 |
| pytorch/torchchat | [UX] We are too quiet about errors - in particular missing HF authentication... | 3 |
| pytorch/torchdistx | Documentation for AnyPrecisionOptimizer | 3 |
| pytorch/torchtune | Add add_end_token to Mistral tokenizer | 3 |
| pytorch/torchtune | implement activation offloading and opt_in_bwd in knowledge_distillation recipes | 3 |
| pytorch/xla | Op info test for nn.functional.leaky_relu .. nn.functional.max_pool1d | 3 |
| pytorch/xla | Op info test for kthvalue | 3 |
| xbmc/addon-check | Correct the mentioned pylint errors to get perfect 10. (missing-docstring) | 3 |
| godotengine/godot-blender-exporter | Blender object with negative scale | 3 |
| huggingface/accelerate | RuntimeError: Tensor on device meta is not on the expected device cuda:0! | 4 |
| huggingface/accelerate | Barebones dataloader to allow for any type of iterable dataloader-like object to be used. Should just handle device placement | 4 |
| huggingface/dataset-viewer | Cache/Queue metrics should not be negative | 4 |
| huggingface/lighteval | [EVAL]: Add more African Benchmarks | 4 |
| huggingface/nanotron | Avoid nested InheritFromOtherOptimizer | 4 |
| huggingface/transformers | train_tokens_per_second is wrong after continuing from checkpoint | 4 |
| huggingface/transformers | OWL-VIT Vision Foundation Model deployment in the edge cases - Need SDPA support for OWL-ViT Model optimization for Edge Deployment | 4 |
| huggingface/transformers | GeneratorExp aren't supported by torch.jit.script when I try to export a previously trained model 'google/vit-base-patch16-224-in21k'. | 4 |
| pandas-dev/pandas | BUG: DataFrame.mean(axis=1) fails but Series.mean works incase of pandas Nullable integer dtype | 4 |
| pandas-dev/pandas-stubs | Type checking with converters= on read_excel | 4 |
| ansible/ansible-documentation | Link to devel in the various roadmaps | 4 |
| ansible/awx | visualizer info field | 4 |
| ansible/awx | Remove some of the eslint-disable comments | 4 |
| ansible/awx | Docker install instructions do not mention how to add awx-logos | 4 |
| ansible/django-ansible-base | Remove model imports in any utils/ modules | 4 |
| tensorflow/graphics | Add ScanNet dataset | 4 |
| pytorch/ao | Quantized Training | 4 |
| pytorch/examples | Please change dcgan to load truncated images. | 4 |
| pytorch/examples | No matching function call error in custom_dataset example | 4 |
| pytorch/examples | Add Siamese Network example | 4 |
| pytorch/pytorch | aten.grid_sampler_3d.default is missing a c-shim implementation, using proxy executor as fallback | 4 |
| pytorch/pytorch | torch.fx.Tracer.record_stack_traces is broken in torch 2.4.0 | 4 |
| pytorch/rl | [Feature Request] Missing ActionScaling and FlattenAction | 4 |
| pytorch/torchcodec | Use TORCH_CHECK instead of throw | 4 |
| pytorch/torchtune | [Docs] Recipe documentation tracker | 4 |
| pytorch/xla | Op info test for pinverse .. put | 4 |
| pytorch/xla | Op info test for nn.functional.avg_pool1d .. nn.functional.bilinear | 4 |
| pytorch/xla | Op info test for linalg.lu_solve | 4 |
| huggingface/accelerate | [Docs] Update low-precision training docs for MS-AMP | 5 |
| huggingface/lighteval | [BUG] custom model docs don't run: missing imports | 5 |
| huggingface/lighteval | Append revision to filepath in --output_dir? | 5 |
| huggingface/nanotron | [Bug] Missing _is_using_mup when resume checkpoint | 5 |
| huggingface/nanotron | [Unit Test] Add unit tests for DistributedTrainer | 5 |
| huggingface/optimum-benchmark | Evaluators for specific tasks | 5 |
| huggingface/transformers | RuntimeError when loading llmcompressor W8A8 quantized model: int8 dtype in weight initialization | 5 |
| pandas-dev/pandas | ENH: Change ExtensionOpsMixin behaviour to not add new operator method if one is already defined on the ExtensionArray class | 5 |
| ansible/awx | adhoc jobs block other jobs from being processed in the queue | 5 |
| ansible/awx | all cluster nodes in development environment have the same UUID | 5 |
| tensorflow/graphics | Not detecting TF Nightly when installing from whl. | 5 |
| tensorflow/similarity | Implement multi-siam for segmentation | 5 |
| pytorch/ao | [low-bit optim] Add COAT optimizer | 5 |
| pytorch/ao | The next tutorials | 5 |
| pytorch/audio | read mp3 file fail | 5 |
| pytorch/pytorch | get_ema_multi_avg_fn() equation is a little confused | 5 |
| pytorch/pytorch | Improve error message for wrong number of arguments in CachingAutotuner | 5 |
| pytorch/pytorch | If mark_dynamic fails (Not all values of RelaxedUnspecConstraint are valid) due to specialization, error message should print where specialization came from | 5 |
| pytorch/pytorch | AOTAutograd: functionalization should emit foreach_copy_() instead of copy_() when possible | 5 |
| pytorch/torchft | Use bucketized model averaging for LocalSGD | 5 |
| pytorch/torchft | [CheckpointServer] use streaming transfers | 5 |
| huggingface/lighteval | [FT] Improve Documentation and Examples | 6 |
| huggingface/transformers | Safetensors files for long-t5-tglobal models fail to load correctly | 6 |
| pandas-dev/pandas | BUG: Series.clip does not work with scalar numpy arrays. | 6 |
| pandas-dev/pandas | BUG: concat_{categorical, compat} leads to erroneous result on non-ns datetime-EA | 6 |
| ansible/ansible-documentation | Misleading statement about "defaults" and "vars" sub-folders of roles | 6 |
| ansible/ansible-navigator | Allow to set ansible-runner artifacts dir ID | 6 |
| ansible/awx | AWX Collection Credential Delete | 6 |
| tensorflow/similarity | Create a colab for single shot to demonstrate how augmenters works | 6 |
| pytorch/ao | Accelerate activation sparsity with activation compression | 6 |
| pytorch/examples | DCGAN C++ warning after PyTorch update | 6 |
| pytorch/executorch | Manual kernel registration to include library names in API | 6 |
| pytorch/executorch | [Android] Add a Runtime.java | 6 |
| pytorch/ignite | Fix FutureWarning: torch.cuda.amp.GradScaler(args...) is deprecated. Please use torch.amp.GradScaler('cuda', args...) instead. | 6 |
| pytorch/pytorch | dynamo (re)compilation issues: shape (1,1), nn.Parameter, mark_dynamic | 6 |
| huggingface/datasets | WMT21 & WMT22 | 7 |
| huggingface/huggingface_hub | [Community event] Translate documentation to your own langage | 7 |
| huggingface/transformers | oom when using adafactor optimizer in deepspeed | 7 |
| pandas-dev/pandas | ERR: numeric_only in reduction operations should disallow passing non-bools | 7 |
| pandas-dev/pandas | startingMonth ignored on non-unitary Quarter periods | 7 |
| tensorflow/similarity | Implement SwAV for self-supervised learning | 7 |
| pytorch/audio | Use non-persistent buffers | 7 |
| pytorch/examples | Why don't we use MSE as a reconstruction loss for VAE ? | 7 |
| pytorch/executorch | Rename executorch cmake target to prim_ops_lib | 7 |
| pytorch/pytorch | [Feature request] torch.export .save/.load could support safetensors and/or weights_only=True | 7 |
| pytorch/pytorch | [BUG][PyTorch 2.0 Export][quant]:get_source_partitions() may return different matches with same input graph | 7 |
| pytorch/pytorch | torch.nn.InstanceNorm2d throws "mixed dtype" error with track_running_stats set to True | 7 |
| pytorch/pytorch | reduce torch.compile default logs | 7 |
| huggingface/diffusers | Make FlaxLMSDiscreteScheduler jittable | 8 |
| huggingface/lighteval | [EVAL] Adding PHARE | 8 |
| huggingface/transformers | Add argument to set number of eval steps in Trainer | 8 |
| pandas-dev/pandas | DOC: NaT - "alias of NaT"; NA - "alias of " | 8 |
| pandas-dev/pandas | DOC: Enforce Numpy Docstring Validation (Parent Issue) | 8 |
| pandas-dev/pandas | BUG: itemsize wrong for date32[day][pyarrow] dtype? | 8 |
| pandas-dev/pandas | BUG: AssertionError: Did not expect new dtype float64 to equal self.dtype float64. Please report a bug | 8 |
| pandas-dev/pandas | BUG: DataFrame.rank does not return EA types when original type was an EADtype | 8 |
| pandas-dev/pandas | DOC: .str.cat output in case of Index object | 8 |
| pandas-dev/pandas | API/BUG: freq retention in value_counts | 8 |
| ansible/awx | Display SAML login icon for /api/login page | 8 |
| pytorch/pytorch | [Inductor] Inference failed with the compiled model with aminmax operator | 8 |
| pytorch/pytorch | [Dynamo] Support tracing through _get_current_dispatch_mode_stack | 8 |
| huggingface/datasets | Identical keywords in build_kwargs and config_kwargs lead to TypeError in load_dataset_builder() | 9 |
| huggingface/diffusers | Support multiple control nets in the StableDiffusionControlNetXSPipeline/StableDiffusionXLControlNetXSPipeline | 9 |
| huggingface/huggingface_hub | Allow dry_run for snapshot_download | 9 |
| huggingface/transformers | Trainer.training_step incorrectly normalizes mean token loss when n_gpu > 1 | 9 |
| pandas-dev/pandas | DOC: Incorrect Source Link | 9 |
| pandas-dev/pandas | Information for new contributors | 9 |
| tensorflow/similarity | Lifted Structured loss | 9 |
| pytorch/pytorch | Make torch.random.get_rng_state take a device argument | 9 |
| pytorch/pytorch | foreach_map enhancements | 9 |
| pytorch/pytorch | Using an empty tensor and torch.int64 or torch.bool for dtype of nanmean() works while a non-empty tensor doesn't work | 9 |
| pytorch/pytorch | Dead link in torch.compile docs | 9 |
| huggingface/diffusers | [Flux ControlNet] Add support for de-distilled models with CFG | 10 |
| huggingface/lighteval | [EVAL] Add TUMLU benchmark | 10 |
| huggingface/transformers | Tapas tokenization Different from Tensorflow Code | 10 |
| pytorch/ao | [MPS] torchao low-bit-precision optim does not expose 'backend' argument to torch.compile | 10 |
| pytorch/executorch | Check tensor's dim order ambiguity in IR verifier | 10 |
| pytorch/pytorch | Docs are little bit outdated for torch logs | 10 |
| pytorch/pytorch | Dump bytecode of resumption frames in tlparse | 10 |
| pytorch/pytorch | Torch Inductor Windows Path Escape Characters | 10 |
| pytorch/pytorch | bmm, topk, cholesky, linalg.norm, max with out variants set causing recompilations in torch.compile | 10 |
| pytorch/pytorch | Expand Tag Set: views & reductions | 10 |
| huggingface/datasets | Consider using "Sequence" instead of "List" | 11 |
| huggingface/transformers | [trainer] figuring out why eval with --fp16_full_eval is 25% slower | 11 |
| pandas-dev/pandas | DOC: Example missing in pandas.DataFrame.to_html | 11 |
| pandas-dev/pandas | DOC: Reindexing behaviour of dataframe column-assignment missing | 11 |
| tensorflow/quantum | [Performance] Boost tfq.convert_to_tensor speed | 11 |
| pytorch/ao | Named Symbol not found (torchchat #1298) | 11 |
| pytorch/data | Improve the warning message regarding local function not supported by pickle | 11 |
| pytorch/pytorch | Device check missing in torch.linalg.solve_triangular leading to hard crash | 11 |
| huggingface/transformers | Verify interpolation of image processors | 12 |
| pytorch/pytorch | torch.export.export() fails on GPU with LSTM model: "Cannot access data pointer of Tensor" | 12 |
| pytorch/pytorch | Obscure error: Expected a value of type 'List[int]' for argument 'sizes' but instead found type 'immutable_list' | 12 |
| huggingface/transformers | Unknown Model (mobilenetv5_300m_enc) when loading Gemma 3n | 13 |
| huggingface/transformers | IsADirectoryError when training with tqdm enabled for trainer | 13 |
| huggingface/transformers | Object Detection Pipeline only outputs first element when batching | 13 |
| pandas-dev/pandas | ENH: set __module__ on top-level public objects | 13 |
| huggingface/diffusers | [Pipeline] AnimateDiff + SparseControl + ControlNet | 14 |
| pandas-dev/pandas | BUG: assert_frame_equal(check_dtype=False) fails when comparing two DFs containing pd.NA that only differ in dtype (object vs Int32) | 14 |
| pytorch/pytorch | Make tlparse able to show a summary of distinct graph breaks | 14 |
| huggingface/datasets | Return the name of the currently loaded file in the load_dataset function. | 15 |
| huggingface/transformers | ConditionalDetrImageProcessor still accepts the deprecated parameter max_size | 15 |
| pandas-dev/pandas | BUG: read_sql no longer works simply with SqlAlchemy selectables and a quick fix | 15 |
| huggingface/transformers | FSDP Torch XLA vs. FSDPv2 (SMPD) Torch XLA checkpoint saving bug | 16 |
| huggingface/transformers | Follow ups to DocumentQuestionAnswering Pipeline | 16 |
| pandas-dev/pandas | DOC: Simplify pandas theme footer | 16 |
| pandas-dev/pandas | BUG: Interpolate over time does not work with Int64 or Float64 | 16 |
| pandas-dev/pandas | Indexing on series where index is output from pd.cut | 16 |
| pytorch/ao | Replace flash_4 with FlexAttention | 17 |
| huggingface/transformers | Update no_trainer scripts to include gradient accumulation | 18 |
| pandas-dev/pandas | DOC: add button to edit on GitHub | 18 |
| pytorch/ao | Add codebook (look up table based) quantization flow in torchao | 18 |
| pytorch/audio | Add Stereo to Mono Convertions | 18 |
| pytorch/executorch | [Request impl] Devtool end-to-end tests | 18 |
| pytorch/torchtune | [Clean up] Move evaluation configs under model directories | 18 |
| huggingface/transformers | Accelerate x Trainer issue tracker: | 19 |
| huggingface/transformers | Improve EncoderDecoderModel docs | 19 |
| fastapi/fastapi | Will FastAPI support QUERY http method? "app.query" | 19 |
| huggingface/diffusers | Expanded init fields in StableDiffusionPipeline cause incompatibilities with many/most inherited pipelines | 20 |
| pandas-dev/pandas | WEB: Multiple "Getting Started" on pandas.io leads to different links | 20 |
| huggingface/peft | Comparison of Different Fine-Tuning Techniques for Conversational AI | 21 |
| pandas-dev/pandas | BUG: CustomBusinessDay not respecting calendar | 21 |
| pandas-dev/pandas | DOC: Single Document For Code Guidelines | 21 |
| huggingface/transformers | Export LayoutLMv2 to onnx | 24 |
| pandas-dev/pandas | BUG: groupby.sum() is inconsistent with df.sum() for large integers | 24 |
| huggingface/transformers | Add support for BLIP and GIT in image-to-text and VQA pipelines | 25 |
| huggingface/optimum | Community contribution - BetterTransformer integration for more models! | 26 |
| huggingface/diffusers | [Community Pipelines] | 27 |
| huggingface/transformers | Getting time offsets of beginning and end of each word in Wav2Vec2 | 27 |
| pandas-dev/pandas | CLN: Use dedup_names in all instances where duplicate column names are renamed | 28 |
| pytorch/ignite | 🥇 Everyone is welcome to contribute 💯 | 30 |
| huggingface/transformers | Trying to add support for GPT2 as decoder in EncoderDecoder model | 31 |
| huggingface/transformers | Community contribution: enable dynamic resolution input for more vision models. | 40 |
| huggingface/transformers | [Community Event] Doc Tests Sprint | 104 |
| huggingface/transformers | [Community contributions] Model cards | 191 |
| Repo | Title | Comments |
|---|
| huggingface/ratchet | Reduce dependencies! 🗡️ | 0 |
| tensorflow/rust | Add Scope::with_kernel_label and with_xla_cluster | 3 |
| tensorflow/rust | Add OpArgDef::default_value and allowed_values | 7 |
| Repo | Title | Comments |
|---|
| godotengine/discourse-theme | Change default green to our own green for the Solved questions | 0 |
| godotengine/discourse-theme | Make tag and tag count more distinct | 0 |
| godotengine/discourse-theme | Categories are not rounded on mobile | 0 |
| godotengine/discourse-theme | Search bar should not cover full width on mobile | 0 |
| godotengine/discourse-theme | Remove the "Sign Up" button from navbar | 2 |
| Repo | Title | Comments |
|---|
| tensorflow/swift-apis | [summary] Better formatting | 0 |
| tensorflow/swift-apis | Lift Tensor.dimensionGathering() out of swift-models and into swift-apis | 0 |
| tensorflow/swift-apis | Investigate size/performance tradeoffs for -sil-inline-generics and -sil-partial-specialization | 0 |
| tensorflow/swift-apis | Update doc comments on non-mutating API to use proper wording. | 0 |
| huggingface/sam2-studio | Model download selector | 1 |
| huggingface/chat-macOS | [Feature Request] System Prompt Support | 2 |
| huggingface/chat-macOS | Asking "what time is it?" will always return the local time of Paris, regardless of your location (⌘R+) | 2 |
| tensorflow/swift-apis | Recursive Neural Networks (structured data/trees) | 3 |
| huggingface/chat-macOS | Add custom LLM API | 5 |
| tensorflow/swift-apis | Implement Recurrent Layers | 5 |
| tensorflow/swift-apis | some mutating tensor operations are not @differentiable | 7 |
| tensorflow/swift-apis | Add derivative tests for layers | 7 |
| tensorflow/swift-apis | Adding Test Cases for Layers in Layer.swift file | 7 |
| tensorflow/swift-apis | Support Advanced Layers | 8 |
| tensorflow/swift-apis | Swifty API discovery via '@available' on raw APIs. | 8 |
| tensorflow/swift-apis | Add preconditions | 10 |
| huggingface/swift-transformers | Support for embedding models (BGE, GTE etc) | 12 |
| tensorflow/swift-apis | Add more optimizers and losses | 16 |
| huggingface/swift-transformers | Convert OpenELM to float16 Core ML | 17 |
| huggingface/sam2-studio | Video support estimated release date? | 21 |
| Repo | Title | Comments |
|---|
| ansible/vscode-ansible | Add option to prefix all shell commands the plugin executes | 0 |
| tensorflow/tfjs-tsne | Implement tensorToDataTexture in the GPU | 0 |
| pytorch/test-infra | Better highlight scheme for force merges | 0 |
| huggingface/chat-ui | Add support for non streaming TGI endpoints | 1 |
| pytorch/probot | deprecation of number for octokit | 1 |
| huggingface/huggingface.js | Maximize button not working properly on Hosted inference API block | 2 |
| huggingface/Mongoku | Auto-link ObjectIds to their referenced elements in the same DB | 2 |
| huggingface/chat-ui | Add option for users to customize search engines in settings page | 3 |
| huggingface/chat-ui | Chrome app icon on macOS | 3 |
| huggingface/chat-ui | Load past conversations using the most recent leaf to determine the visible conversation tree. | 3 |
| huggingface/chat-ui | System prompt not taken into account when web browsing. | 3 |
| ansible/vscode-ansible | Option to completely disable autocomplete | 3 |
| huggingface/chat-ui | Explore the possibility for chat-ui to use OpenAI assistants API structure. | 4 |
| pytorch/test-infra | A better HUD menu for PyTorch repos family | 4 |
| huggingface/chat-ui | Disabling Reasoning Summary as an Env Option | 5 |
| huggingface/chat-ui | Support text-generation-webui | 5 |
| pytorch/test-infra | Mergebot merged a PR even though a review had been re-requested | 5 |
| huggingface/chat-ui | Have embedded prompts configurable in private secrets | 7 |
| huggingface/huggingface.js | Tracking integration for Video Classification | 8 |
| huggingface/chat-ui | Current Date Issue | 11 |
| Repo | Title | Comments |
|---|
| huggingface/awesome-huggingface | [hacktoberfest] Hugging Face Collections Hacktoberfest challenge | 0 |
| tensorflow/mlir | Implement support for qualified symbols | 0 |
| tensorflow/mlir | Consider trip count in loop invariant code motion | 4 |
| tensorflow/mlir | [LLVM] Implement missing LLVM IR instructions | 11 |
| Repo | Title | Comments |
|---|
| godotengine/godot-docs | Add C# examples for custom BBCode | 0 |
| godotengine/godot-docs | Curve2D.get_closest_point() incorrectly documented | 0 |
| godotengine/godot-docs | Physics2DDirectSpaceState does not specify local or global coordinates | 0 |
| godotengine/godot-docs | Viewport "use_hdr_2d" documentation point unclear | 1 |
| godotengine/godot-docs | MultiMesh resource buffer description | 1 |
| godotengine/godot-docs | Parallax2D property descriptions do not specify units | 1 |
| godotengine/godot-docs | Specify the coordinate system for the 'force' parameter in Rigidbody apply_force methods | 1 |
| godotengine/godot-docs | make_canvas_position_local | 1 |
| godotengine/godot-docs | Inconsistent code example between PhysicsShapeQueryParameters3D and PhysicsServer3D | 1 |
| godotengine/godot-docs | Rewording/amendment for Camera2D.align () function. | 1 |
| godotengine/godot-docs | AudioServer Documentation does not specify units of time | 1 |
| godotengine/godot-docs | Add clarification about what CharacterBody is used for | 2 |
| godotengine/godot-docs | PhysicsDirectSpaceState2D.get_rest_info claims not to use query motion but it does | 4 |
| godotengine/godot-docs | Control class page doesn't mention ".release_focus()" in intro | 4 |
| godotengine/godot-docs | WorkerThreadPool docs do not specify how Thread count is set | 4 |
Amélioration de la documentation.