# Uncategorized **URL:** https://discuss.pytorch.org/c/uncategorized/1.md [Latest](https://discuss.pytorch.org/latest.md) · [Categories](https://discuss.pytorch.org/categories.md) --- ## [Optimization problem](https://discuss.pytorch.org/t/optimization-problem/225631)
**Author:** [@Ghostv](https://discuss.pytorch.org/u/Ghostv)\ **Replies:** 1\ **Last updated:** [October 24, 2026, 4:29am UTC](https://discuss.pytorch.org/t/optimization-problem/225631 "2026-10-24T04:29:19Z")
Hi! I’m totally new to using torch for solving optimization problems, but for what i understand, i can minimize a loss function after specifying which variables the optimizer can modify. Now, i wanna ask if the kind-of… --- ## [\[Project\] EmoLLM Organiztion: Researching Affective State Transitions & Empathetic Machine Alignment](https://discuss.pytorch.org/t/project-emollm-organiztion-researching-affective-state-transitions-empathetic-machine-alignment/225594)
**Author:** [@sundaram](https://discuss.pytorch.org/u/sundaram)\ **Replies:** 9\ **Last updated:** [October 24, 2026, 12:59am UTC](https://discuss.pytorch.org/t/project-emollm-organiztion-researching-affective-state-transitions-empathetic-machine-alignment/225594 "2026-10-24T00:59:19Z")
Hi PyTorch Community, I wanted to share a research project I’m currently building called affective-lab—a repository focused on affective computing and dynamic emotion tracking between human inputs and machine outputs. M… --- ## [Is there an implementation of edit distance to calculate wer?](https://discuss.pytorch.org/t/is-there-an-implementation-of-edit-distance-to-calculate-wer/71840)
**Author:** [@josipovic](https://discuss.pytorch.org/u/josipovic)\ **Replies:** 4\ **Last updated:** [October 23, 2026, 4:37pm UTC](https://discuss.pytorch.org/t/is-there-an-implementation-of-edit-distance-to-calculate-wer/71840 "2026-10-23T16:37:15Z")
Is there edit\_distance function in pytorch to calculate wer/ler? Its levenshtein distance. There is a pip package https://pypi.org/project/torch-edit-distance/ but its a bit buggy. If there is none in pytorch, why? chee… --- ## [Custom connections in neural network layers](https://discuss.pytorch.org/t/custom-connections-in-neural-network-layers/3027)
**Author:** [@theQmech](https://discuss.pytorch.org/u/theQmech)\ **Replies:** 20\ **Last updated:** [October 23, 2026, 6:30am UTC](https://discuss.pytorch.org/t/custom-connections-in-neural-network-layers/3027 "2026-10-23T06:30:11Z")
At the moment, I’m experimenting with defining custom sparse connections between two fully connected layers of a neural network. To accomplish this, right now I’m modifying nn.Linear(in\_features, out\_features) to nn.… --- ## [Exact source provenance for libgomp.so.1 in PyTorch 2.8.0+cu128 Linux wheel](https://discuss.pytorch.org/t/exact-source-provenance-for-libgomp-so-1-in-pytorch-2-8-0-cu128-linux-wheel/225609)
**Author:** [@akowalski42](https://discuss.pytorch.org/u/akowalski42)\ **Replies:** 0\ **Last updated:** [October 22, 2026, 4:57pm UTC](https://discuss.pytorch.org/t/exact-source-provenance-for-libgomp-so-1-in-pytorch-2-8-0-cu128-linux-wheel/225609 "2026-10-22T16:57:44Z")
Could the Linux wheel release-build custodians help identify the source and build materials actually used for the bundled libgomp.so.1 below? We need these references to assemble a corresponding-source supplement for thi… --- ## [Good Starter Projects](https://discuss.pytorch.org/t/good-starter-projects/225586)
**Author:** [@drezaptmp](https://discuss.pytorch.org/u/drezaptmp)\ **Replies:** 0\ **Last updated:** [October 20, 2026, 10:28pm UTC](https://discuss.pytorch.org/t/good-starter-projects/225586 "2026-10-20T22:28:06Z")
Hi, No specific questions but looking for some open source dev and wanted to introduce myself. I posted on Github so the devs can get a closer look at it, but this might capture a wider audience. I’m not a beginning dev… --- ## [Agentic RAG vs. standard RAG for healthcare: is the extra complexity worth it?](https://discuss.pytorch.org/t/agentic-rag-vs-standard-rag-for-healthcare-is-the-extra-complexity-worth-it/225584)
**Author:** [@antonyfelix7841](https://discuss.pytorch.org/u/antonyfelix7841)\ **Replies:** 0\ **Last updated:** [October 20, 2026, 12:36pm UTC](https://discuss.pytorch.org/t/agentic-rag-vs-standard-rag-for-healthcare-is-the-extra-complexity-worth-it/225584 "2026-10-20T12:36:08Z")
Hi everyone, I’m learning about AI agents and RAG, and I’m trying to understand how they work in a high-stakes field like healthcare, where wrong answers and patient data privacy are serious concerns. Standard RAG retr… --- ## [Request authoritative wheel sizes and release provenance for PyTorch 2.12.1+cu130](https://discuss.pytorch.org/t/request-authoritative-wheel-sizes-and-release-provenance-for-pytorch-2-12-1-cu130/225563)
**Author:** [@ptmedsync](https://discuss.pytorch.org/u/ptmedsync)\ **Replies:** 0\ **Last updated:** [October 16, 2026, 2:24pm UTC](https://discuss.pytorch.org/t/request-authoritative-wheel-sizes-and-release-provenance-for-pytorch-2-12-1-cu130/225563 "2026-10-16T14:24:44Z")
Hello PyTorch Release Engineering Team, We are validating a pinned, offline PyTorch runtime and need authoritative release records for three published torch-2.12.1+cu130 wheels. Please provide the producer-recorded byt… --- ## [\[ROCm\]\[MI200\] linear\_cross\_entropy recomputed logits use backward-only FP16 alternate GEMM path](https://discuss.pytorch.org/t/rocm-mi200-linear-cross-entropy-recomputed-logits-use-backward-only-fp16-alternate-gemm-path/225415)
**Author:** [@dhuang3-amd](https://discuss.pytorch.org/u/dhuang3-amd)\ **Replies:** 1\ **Last updated:** [October 5, 2026, 1:33pm UTC](https://discuss.pytorch.org/t/rocm-mi200-linear-cross-entropy-recomputed-logits-use-backward-only-fp16-alternate-gemm-path/225415 "2026-10-05T13:33:52Z")
Description I investigated three fp16 linear\_cross\_entropy test failures on gfx90a / MI200 originally reported in ROCm/TheRock#6993 and was able to reproduce them and trace the numerical behavior further. The failures a… --- ## [\[Project Announcement\] numgraph: A High-Performance C++ Numerical Core for Learnable Graphs (Looking for Maintainers!)](https://discuss.pytorch.org/t/project-announcement-numgraph-a-high-performance-c-numerical-core-for-learnable-graphs-looking-for-maintainers/225463)
**Author:** [@sundaram](https://discuss.pytorch.org/u/sundaram)\ **Replies:** 2\ **Last updated:** [September 30, 2026, 3:51pm UTC](https://discuss.pytorch.org/t/project-announcement-numgraph-a-high-performance-c-numerical-core-for-learnable-graphs-looking-for-maintainers/225463 "2026-09-30T15:51:05Z")
Hi everyone, I am building numgraph, a lightweight and high-performance numerical core for learnable graphs, designed with a C++ backend and clean Python bindings to handle complex graph-based tensor operations and deep… --- ## [Possible resources leak in void ProcessGroupNCCL::shutdown()](https://discuss.pytorch.org/t/possible-resources-leak-in-void-processgroupnccl-shutdown/225317)
**Author:** [@zk\_z](https://discuss.pytorch.org/u/zk_z)\ **Replies:** 1\ **Last updated:** [September 30, 2026, 9:24am UTC](https://discuss.pytorch.org/t/possible-resources-leak-in-void-processgroupnccl-shutdown/225317 "2026-09-30T09:24:02Z")
While reviewing the NCCL symmetric-memory cleanup path, I noticed a possible cleanup and lifetime issue in ProcessGroupNCCL::shutdown(). This is currently a code-review question; ProcessGroupNCCL::shutdown() may destroy… --- ## [DataLoader shared-memory unlink race on Python 3.14 only after midnight (Ubuntu 22.04, PyTorch 2.12.0+cu130)](https://discuss.pytorch.org/t/dataloader-shared-memory-unlink-race-on-python-3-14-only-after-midnight-ubuntu-22-04-pytorch-2-12-0-cu130/225460)
**Author:** [@evenrose](https://discuss.pytorch.org/u/evenrose)\ **Replies:** 0\ **Last updated:** [September 28, 2026, 1:08pm UTC](https://discuss.pytorch.org/t/dataloader-shared-memory-unlink-race-on-python-3-14-only-after-midnight-ubuntu-22-04-pytorch-2-12-0-cu130/225460 "2026-09-28T13:08:04Z")
Environment: PyTorch 2.12.0+cu130, Python 3.14, Ubuntu 22.04, CUDA 13.0. In a standard training loop using DataLoader with num\_workers \> 0, training occasionally crashes with RuntimeError: could not unlink the shared mem… --- ## [torch.distributions.kumaraswamy.Kumaraswamy generates samples outside its support](https://discuss.pytorch.org/t/torch-distributions-kumaraswamy-kumaraswamy-generates-samples-outside-its-support/173404)
**Author:** [@kiranchari](https://discuss.pytorch.org/u/kiranchari)\ **Replies:** 12\ **Last updated:** [September 24, 2026, 2:13pm UTC](https://discuss.pytorch.org/t/torch-distributions-kumaraswamy-kumaraswamy-generates-samples-outside-its-support/173404 "2026-09-24T14:13:57Z")
The Kumaraswamy distribution is defined over the open interval (0,1) but torch.distributions.kumaraswamy.Kumaraswamy can generate samples that equal 1. This causes “nan” outputs when computing the log\_prob of such sample… --- ## [Comparing raw MRI vs FreeSurfer-derived features for multimodal fusion with a small dataset](https://discuss.pytorch.org/t/comparing-raw-mri-vs-freesurfer-derived-features-for-multimodal-fusion-with-a-small-dataset/225429)
**Author:** [@Reynard-the-Fox](https://discuss.pytorch.org/u/Reynard-the-Fox)\ **Replies:** 0\ **Last updated:** [September 17, 2026, 5:57pm UTC](https://discuss.pytorch.org/t/comparing-raw-mri-vs-freesurfer-derived-features-for-multimodal-fusion-with-a-small-dataset/225429 "2026-09-17T17:57:07Z")
Hi, I’m an MSc student and pretty new to medical imaging/deep learning. I’m working on an Alzheimer’s classification project using the ANMerge dataset. It has around 1,700 participants overall, but only around 450 have M… --- ## [Official wheel downloads return HTTP 403](https://discuss.pytorch.org/t/official-wheel-downloads-return-http-403/225404)
**Author:** [@Jesse\_Yang](https://discuss.pytorch.org/u/Jesse_Yang)\ **Replies:** 3\ **Last updated:** [September 16, 2026, 7:59pm UTC](https://discuss.pytorch.org/t/official-wheel-downloads-return-http-403/225404 "2026-09-16T19:59:14Z")
I’m downloading—but not installing—Linux CPython 3.12 wheels for a separate machine, using Python 3.14 on Windows 11. These four packages all return HTTP 403 from download-r2.pytorch.org: torch 2.9.1+cu130 torchaudio … --- ## [\[Showcase\] pyforker: A python framework for \[core functionality\]](https://discuss.pytorch.org/t/showcase-pyforker-a-python-framework-for-core-functionality/225412)
**Author:** [@sundaram](https://discuss.pytorch.org/u/sundaram)\ **Replies:** 1\ **Last updated:** [September 14, 2026, 1:04pm UTC](https://discuss.pytorch.org/t/showcase-pyforker-a-python-framework-for-core-functionality/225412 "2026-09-14T13:04:11Z")
Hi everyone! 👋 I recently published \`pyforker\` on PyPI (https://pypi.org/project/pyforker/) to help streamline \[explain core problem: e.g., running parallel process forks / task isolation\] in Python workflows. ### 🌟 Ke… --- ## [WebGPU visual PyTorch debugger](https://discuss.pytorch.org/t/webgpu-visual-pytorch-debugger/225211)
**Author:** [@Piotr\_Gryko](https://discuss.pytorch.org/u/Piotr_Gryko)\ **Replies:** 2\ **Last updated:** [September 11, 2026, 3:46am UTC](https://discuss.pytorch.org/t/webgpu-visual-pytorch-debugger/225211 "2026-09-11T03:46:28Z")
Hi everyone, I have built a 3d real time PyTorch debugger. It works directly from python and runs in your browser. I reverse engineered most of the basic PyTorch modules. All of them are rendered in 3d with real time int… --- ## [\[Question\] What is the relationship between PyTorch Symmetric Memory and Torchcomms?](https://discuss.pytorch.org/t/question-what-is-the-relationship-between-pytorch-symmetric-memory-and-torchcomms/225394)
**Author:** [@guitang001](https://discuss.pytorch.org/u/guitang001)\ **Replies:** 0\ **Last updated:** [September 9, 2026, 11:56am UTC](https://discuss.pytorch.org/t/question-what-is-the-relationship-between-pytorch-symmetric-memory-and-torchcomms/225394 "2026-09-09T11:56:52Z")
Are they redundant efforts?As mentioned in the title, I’m exploring the memory management techniques in Torchcomms. However, I noticed that PyTorch already has a Symmetric Memory mechanism, so I’m unclear about the rela… --- ## [\[Maxpooling\] what should the indices be when there're multiple max values?](https://discuss.pytorch.org/t/maxpooling-what-should-the-indices-be-when-therere-multiple-max-values/162390)
**Author:** [@ryanbty](https://discuss.pytorch.org/u/ryanbty)\ **Replies:** 4\ **Last updated:** [September 9, 2026, 7:29am UTC](https://discuss.pytorch.org/t/maxpooling-what-should-the-indices-be-when-therere-multiple-max-values/162390 "2026-09-09T07:29:05Z")
What to do when there are multiple values in the kernel that equals to the max? For example, for these values when we use torch.argmax : torch.tensor(\[\[1., 1.\], \[1., 1.\]\]) The max is simply 1. What should max in… --- ## [Need help with an error](https://discuss.pytorch.org/t/need-help-with-an-error/223917)
**Author:** [@vs-cb](https://discuss.pytorch.org/u/vs-cb)\ **Replies:** 7\ **Last updated:** [September 7, 2026, 1:01pm UTC](https://discuss.pytorch.org/t/need-help-with-an-error/223917 "2026-09-07T13:01:25Z")
So my application uses easyocr and it has a dependency on pytorch. I’m getting the following error when I run my application as an exe. OSError: \[WinError 1114\] A dynamic link library (DLL) initialization routine failed… --- ## [What does “normal random sampling” mean in LLM text generation?](https://discuss.pytorch.org/t/what-does-normal-random-sampling-mean-in-llm-text-generation/225384)
**Author:** [@nono345](https://discuss.pytorch.org/u/nono345)\ **Replies:** 0\ **Last updated:** [September 7, 2026, 1:00pm UTC](https://discuss.pytorch.org/t/what-does-normal-random-sampling-mean-in-llm-text-generation/225384 "2026-09-07T13:00:57Z")
I am working on research involving the numerical behavior of LLM token sampling, and I need to reproduce the ordinary stochastic next-token sampling process as faithfully as possible. At each decoding step, I obtain the… --- ## [TypeError: Object of type int32 is not JSON serializable](https://discuss.pytorch.org/t/typeerror-object-of-type-int32-is-not-json-serializable/169762)
**Author:** [@mattpats](https://discuss.pytorch.org/u/mattpats)\ **Replies:** 2\ **Last updated:** [September 5, 2026, 9:26am UTC](https://discuss.pytorch.org/t/typeerror-object-of-type-int32-is-not-json-serializable/169762 "2026-09-05T09:26:03Z")
I am training a TabNetClassifier. My code is largely borrowed from: ‘tabnet/census\_example.ipynb at develop · dreamquark-ai/tabnet · GitHub’ Everything is working fine until I try to save the model. When trying to save… --- ## [My index is out of bounds for some graph in gcn model](https://discuss.pytorch.org/t/my-index-is-out-of-bounds-for-some-graph-in-gcn-model/225319)
**Author:** [@mihai\_t](https://discuss.pytorch.org/u/mihai_t)\ **Replies:** 1\ **Last updated:** [September 1, 2026, 10:37am UTC](https://discuss.pytorch.org/t/my-index-is-out-of-bounds-for-some-graph-in-gcn-model/225319 "2026-09-01T10:37:58Z")
I am trying to do a graph convolution network with forward-forward propagation. When I am running the model, I get an error saying that my index is out of bounds, but this happens after the model works fine for some exam… --- ## [SNN Energy Forecasting Peak Underprediction](https://discuss.pytorch.org/t/snn-energy-forecasting-peak-underprediction/225362)
**Author:** [@Ahaan\_Desai](https://discuss.pytorch.org/u/Ahaan_Desai)\ **Replies:** 0\ **Last updated:** [September 1, 2026, 8:14am UTC](https://discuss.pytorch.org/t/snn-energy-forecasting-peak-underprediction/225362 "2026-09-01T08:14:28Z")
I’m working on an energy forecasting project using a Spiking Neural Network (SNN), and I’m trying to understand why my model is severely underpredicting high-load events. Task: Dataset: UK Electrical Load / House 4 Da… --- ## [Transfer learning in semantic segmentation on Cityscapes dataset](https://discuss.pytorch.org/t/transfer-learning-in-semantic-segmentation-on-cityscapes-dataset/180358)
**Author:** [@Ves](https://discuss.pytorch.org/u/Ves)\ **Replies:** 1\ **Last updated:** [August 31, 2026, 1:10pm UTC](https://discuss.pytorch.org/t/transfer-learning-in-semantic-segmentation-on-cityscapes-dataset/180358 "2026-08-31T13:10:29Z")
Hello there. I am totally new to Semantic Segmentation, and my current project is observing effects of domain shifting in Semantic Segmentation. I was thinking of using Cityscapes as my real data set and GTA alongside S… --- ## [CUDA 13.3 Update No Compatibility](https://discuss.pytorch.org/t/cuda-13-3-update-no-compatibility/225338)
**Author:** [@Isaac1](https://discuss.pytorch.org/u/Isaac1)\ **Replies:** 3\ **Last updated:** [August 31, 2026, 3:20am UTC](https://discuss.pytorch.org/t/cuda-13-3-update-no-compatibility/225338 "2026-08-31T03:20:48Z")
I downloaded CUDA 13.3 as there new update but now I cant run PyTorch cause the latest version you support is CUDA 13.2. Could you please update this? --- ## [Request to join PyTorch Slack with a gr-oss.io or gresearch.co.uk email](https://discuss.pytorch.org/t/request-to-join-pytorch-slack-with-a-gr-oss-io-or-gresearch-co-uk-email/225342)
**Author:** [@tolleybot](https://discuss.pytorch.org/u/tolleybot)\ **Replies:** 0\ **Last updated:** [August 29, 2026, 6:51pm UTC](https://discuss.pytorch.org/t/request-to-join-pytorch-slack-with-a-gr-oss-io-or-gresearch-co-uk-email/225342 "2026-08-29T18:51:19Z")
Hi PyTorch Team, I contribute to PyTorch as part of Insight Softmax, working with G-Research on open source engineering. Five merged PRs since October 2024 across inductor, export serialization, and core tensor operatio… --- ## [Pytorch docathon merch](https://discuss.pytorch.org/t/pytorch-docathon-merch/225341)
**Author:** [@Jonathan\_Coletti](https://discuss.pytorch.org/u/Jonathan_Coletti)\ **Replies:** 0\ **Last updated:** [August 29, 2026, 6:50pm UTC](https://discuss.pytorch.org/t/pytorch-docathon-merch/225341 "2026-08-29T18:50:15Z")
Thank you Pytorch for sending this awesome merch. I thought I would share it with everyone --- ## [Request to join PyTorch Slack with an apache.org email](https://discuss.pytorch.org/t/request-to-join-pytorch-slack-with-an-apache-org-email/225339)
**Author:** [@400Ping](https://discuss.pytorch.org/u/400Ping)\ **Replies:** 0\ **Last updated:** [August 28, 2026, 11:06pm UTC](https://discuss.pytorch.org/t/request-to-join-pytorch-slack-with-an-apache-org-email/225339 "2026-08-28T23:06:07Z")
Hi PyTorch team, I’m a PMC member at Apache Mahout. The module I’m working on in Mahout uses PyTorch as part of our development and performance validation workflows. I’d like to join the PyTorch Slack so I can better p… --- ## [Built a tiny tool after my own validation numbers kept lying to me (metrictrust)](https://discuss.pytorch.org/t/built-a-tiny-tool-after-my-own-validation-numbers-kept-lying-to-me-metrictrust/225337)
**Author:** [@davidtoky](https://discuss.pytorch.org/u/davidtoky)\ **Replies:** 0\ **Last updated:** [August 27, 2026, 7:59am UTC](https://discuss.pytorch.org/t/built-a-tiny-tool-after-my-own-validation-numbers-kept-lying-to-me-metrictrust/225337 "2026-08-27T07:59:51Z")
Hey everyone, while fine-tuning small models, I kept running into the same trap three times over: the model hit its best score mid-training, and I ended up keeping a worse final checkpoint train accuracy was much highe… [Next page](https://discuss.pytorch.org/c/uncategorized/1.md?page=1)