# projects **URL:** https://discuss.pytorch.org/c/projects/33.md?page=1 [Latest](https://discuss.pytorch.org/latest.md) · [Categories](https://discuss.pytorch.org/categories.md) **Page:** 2 --- ## [Pytorch Symbolic: an equivalent of Keras Functional API](https://discuss.pytorch.org/t/pytorch-symbolic-an-equivalent-of-keras-functional-api/165395)
**Author:** [@gahaalt](https://discuss.pytorch.org/u/gahaalt)\ **Replies:** 3\ **Last updated:** [January 18, 2026, 4:24pm UTC](https://discuss.pytorch.org/t/pytorch-symbolic-an-equivalent-of-keras-functional-api/165395 "2026-01-18T16:24:37Z")
Hello! I just hit 1.0.0 version of a library I’ve been developing for the past months as a side project. Pytorch Symbolic A library that aims to provide a concise API for neural network creation in PyTorch. The API and… --- ## [User embeddings based on sequence of events](https://discuss.pytorch.org/t/user-embeddings-based-on-sequence-of-events/224083)
**Author:** [@Ihor\_Bobak](https://discuss.pytorch.org/u/Ihor_Bobak)\ **Replies:** 0\ **Last updated:** [December 2, 2025, 9:59pm UTC](https://discuss.pytorch.org/t/user-embeddings-based-on-sequence-of-events/224083 "2025-12-02T21:59:48Z")
Hi everyone. I work in ad-tech, and we’ve started investigating user embeddings using a Sequence-of-Events (SoE) approach - where embeddings are built not on aggregated features, but directly from raw user events. We’ve… --- ## [QuaTorch - Quaternions in Pytorch](https://discuss.pytorch.org/t/quatorch-quaternions-in-pytorch/223567)
**Author:** [@lucasegidio1](https://discuss.pytorch.org/u/lucasegidio1)\ **Replies:** 4\ **Last updated:** [November 4, 2025, 3:28am UTC](https://discuss.pytorch.org/t/quatorch-quaternions-in-pytorch/223567 "2025-11-04T03:28:03Z")
Hi all, Please allow me to advertise this project I started: QuaTorch Try it out with pip install quatorch The idea is to have lightweight package providing a subclass of torch.Tensor , namely, quatorch.Quaternion th… --- ## [Need advice (or maybe a mentor) for building an ML pipeline for sales forecasting](https://discuss.pytorch.org/t/need-advice-or-maybe-a-mentor-for-building-an-ml-pipeline-for-sales-forecasting/223716)
**Author:** [@Daniil\_Bystrov](https://discuss.pytorch.org/u/Daniil_Bystrov)\ **Replies:** 2\ **Last updated:** [November 2, 2025, 12:52pm UTC](https://discuss.pytorch.org/t/need-advice-or-maybe-a-mentor-for-building-an-ml-pipeline-for-sales-forecasting/223716 "2025-11-02T12:52:50Z")
Hey everyone :waving\_hand: I’m currently working on a project where I need to forecast meat sales per product (article) — sounds simple, but it’s getting a bit messy. Right now, I have an Excel sheet with 4 product … --- ## [Training Loop made easy](https://discuss.pytorch.org/t/training-loop-made-easy/223496)
**Author:** [@mobadara](https://discuss.pytorch.org/u/mobadara)\ **Replies:** 0\ **Last updated:** [October 7, 2025, 8:07am UTC](https://discuss.pytorch.org/t/training-loop-made-easy/223496 "2025-10-07T08:07:10Z")
:fire: Introducing Torchflow — A Lightweight Library to Streamline PyTorch Training, Tuning, and Callbacks Hello PyTorch Community :waving\_hand:, I’m excited to share my first open-source project, Torchflow, now availa… --- ## [WhyTorch - a visual explainer for PyTorch functions](https://discuss.pytorch.org/t/whytorch-a-visual-explainer-for-pytorch-functions/223460)
**Author:** [@Adam\_Allevato](https://discuss.pytorch.org/u/Adam_Allevato)\ **Replies:** 3\ **Last updated:** [October 6, 2025, 10:37am UTC](https://discuss.pytorch.org/t/whytorch-a-visual-explainer-for-pytorch-functions/223460 "2025-10-06T10:37:45Z")
I just built WhyTorch, an open source website I built to explain PyTorch functions. It makes tricky functions like torch.gather and torch.scatter more intuitive by showing element-level relationships between inputs and o… --- ## [Need some advice on 1D CNN trained on synthetic spectra to predict the frequency of a single minimum on real signal](https://discuss.pytorch.org/t/need-some-advice-on-1d-cnn-trained-on-synthetic-spectra-to-predict-the-frequency-of-a-single-minimum-on-real-signal/223326)
**Author:** [@Marklee01](https://discuss.pytorch.org/u/Marklee01)\ **Replies:** 0\ **Last updated:** [September 26, 2025, 2:43am UTC](https://discuss.pytorch.org/t/need-some-advice-on-1d-cnn-trained-on-synthetic-spectra-to-predict-the-frequency-of-a-single-minimum-on-real-signal/223326 "2025-09-26T02:43:42Z")
Hi. I’m an undergraduate studying plasma, trying to apply deep learning to my research. I don’t have local ML support, so I had to study it all by myself and get some help from AI tools. I ’m seeking practical advice fro… --- ## [THOAD, Arbitrary Order Automatic Differentiation for PyTorch](https://discuss.pytorch.org/t/thoad-arbitrary-order-automatic-differentiation-for-pytorch/222851)
**Author:** [@mntsx](https://discuss.pytorch.org/u/mntsx)\ **Replies:** 0\ **Last updated:** [September 3, 2025, 6:20pm UTC](https://discuss.pytorch.org/t/thoad-arbitrary-order-automatic-differentiation-for-pytorch/222851 "2025-09-03T18:20:43Z")
Presenting the package I’m excited to finally release the first version of thoad (short for PyTorch High Order Automatic Differentiation), a Python only library that computes arbitrary order partial derivatives directly … --- ## [Guiding Image Generation in Diffusion Models with PyTorch](https://discuss.pytorch.org/t/guiding-image-generation-in-diffusion-models-with-pytorch/222779)
**Author:** [@sid-ls](https://discuss.pytorch.org/u/sid-ls)\ **Replies:** 0\ **Last updated:** [August 31, 2025, 9:26am UTC](https://discuss.pytorch.org/t/guiding-image-generation-in-diffusion-models-with-pytorch/222779 "2025-08-31T09:26:21Z")
I just open-sourced an introductory-friendly PyTorch implementation for controlling image generation in diffusion models. It’s built on CASteer, which lets us apply pre-computed steering vectors to guide models toward ge… --- ## [How do you structure a stateful LSTM with PyTorch?](https://discuss.pytorch.org/t/how-do-you-structure-a-stateful-lstm-with-pytorch/189358)
**Author:** [@zezimabig](https://discuss.pytorch.org/u/zezimabig)\ **Replies:** 4\ **Last updated:** [August 30, 2025, 4:36pm UTC](https://discuss.pytorch.org/t/how-do-you-structure-a-stateful-lstm-with-pytorch/189358 "2025-08-30T16:36:12Z")
Dear PyTorch community, I apologise for the uncategorizing of this particular question as I could not find a forecasting category. Basically, I would like to find out if anyone has ever tried to define a stateful LSTM … --- ## [Jupyter notebook repository for self-learning](https://discuss.pytorch.org/t/jupyter-notebook-repository-for-self-learning/222622)
**Author:** [@vdw](https://discuss.pytorch.org/u/vdw)\ **Replies:** 0\ **Last updated:** [August 24, 2025, 11:28am UTC](https://discuss.pytorch.org/t/jupyter-notebook-repository-for-self-learning/222622 "2025-08-24T11:28:37Z")
I hope this fits the posting guidelines here: For self-learners in the wider area of AI/ML/DL/NLP but also data science in general, we are building a repository of Jupyter notebooks. The repository can be seen as an open… --- ## [Package to interactively visualize large Pytorch models in notebooks - torchvista](https://discuss.pytorch.org/t/package-to-interactively-visualize-large-pytorch-models-in-notebooks-torchvista/221845)
**Author:** [@dev-table](https://discuss.pytorch.org/u/dev-table)\ **Replies:** 2\ **Last updated:** [August 15, 2025, 11:31pm UTC](https://discuss.pytorch.org/t/package-to-interactively-visualize-large-pytorch-models-in-notebooks-torchvista/221845 "2025-08-15T23:31:48Z")
Hi, I want to share a new open source package I’ve created to interactively visualize Pytorch models in notebooks (like Jupyter, Google Colab etc). It is called torchvista. Github page With one line of code, you can t… --- ## [Blace.ai - new C++ Inference SDK & Model Hub](https://discuss.pytorch.org/t/blace-ai-new-c-inference-sdk-model-hub/222361)
**Author:** [@spiegelball](https://discuss.pytorch.org/u/spiegelball)\ **Replies:** 0\ **Last updated:** [August 14, 2025, 3:43pm UTC](https://discuss.pytorch.org/t/blace-ai-new-c-inference-sdk-model-hub/222361 "2025-08-14T15:43:03Z")
Hey, I’ve been working with C++ and LibTorch for a while, and kept running into the same recurring issues — figuring out the correct build for each OS, switching between CUDA, MPS, and CPU backends, and dealing with GPU… --- ## [How to address pretrained facenet overfitting for facial verification?](https://discuss.pytorch.org/t/how-to-address-pretrained-facenet-overfitting-for-facial-verification/221844)
**Author:** [@Jayasuryan](https://discuss.pytorch.org/u/Jayasuryan)\ **Replies:** 0\ **Last updated:** [July 27, 2025, 3:47am UTC](https://discuss.pytorch.org/t/how-to-address-pretrained-facenet-overfitting-for-facial-verification/221844 "2025-07-27T03:47:21Z")
Hello everyone, I’m currently working on a building a facial verification system using facenet-pytorch. I would like to ask for some guidance with this project as I have observed my model was overfitting. I will be expl… --- ## [Torchzero - modular optimization library](https://discuss.pytorch.org/t/torchzero-modular-optimization-library/221520)
**Author:** [@qq-me](https://discuss.pytorch.org/u/qq-me)\ **Replies:** 0\ **Last updated:** [July 14, 2025, 7:56pm UTC](https://discuss.pytorch.org/t/torchzero-modular-optimization-library/221520 "2025-07-14T19:56:41Z")
I made a library about optimization GitHub - inikishev/torchzero: Modular optimization library for PyTorch. The basic idea is that everything is a module and different modules can be combined, a basic example of Cautiou… --- ## [Where to get the version string of Lightning runs?](https://discuss.pytorch.org/t/where-to-get-the-version-string-of-lightning-runs/220278)
**Author:** [@evilroach](https://discuss.pytorch.org/u/evilroach)\ **Replies:** 4\ **Last updated:** [May 24, 2025, 3:40am UTC](https://discuss.pytorch.org/t/where-to-get-the-version-string-of-lightning-runs/220278 "2025-05-24T03:40:02Z")
I’m using Lightning to train my model. Lightning has a nice feature i.e. every time it runs, it creates a sub-dir in ./lightning\_logs, and nams them like version\_0, version\_1, version\_2… I want to utilize these strings… --- ## [Custom Fold and Unfold](https://discuss.pytorch.org/t/custom-fold-and-unfold/219793)
**Author:** [@AnomUser8765](https://discuss.pytorch.org/u/AnomUser8765)\ **Replies:** 0\ **Last updated:** [May 6, 2025, 6:56am UTC](https://discuss.pytorch.org/t/custom-fold-and-unfold/219793 "2025-05-06T06:56:40Z")
def flatten(inputs: torch.Tensor, start\_dim: int = 0, end\_dim: int = -1) -\> torch.Tensor: """ Flattens a tensor from start\_dim to the end. """ # TODO n\_dim = len(inputs.shape) if start\_dim \> n\_dim… --- ## [Improved PyTorch Models in Minutes with Perforated Backpropagation](https://discuss.pytorch.org/t/improved-pytorch-models-in-minutes-with-perforated-backpropagation/219592)
**Author:** [@PerforatedAI](https://discuss.pytorch.org/u/PerforatedAI)\ **Replies:** 0\ **Last updated:** [April 29, 2025, 1:05pm UTC](https://discuss.pytorch.org/t/improved-pytorch-models-in-minutes-with-perforated-backpropagation/219592 "2025-04-29T13:05:13Z")
I’ve developed a new optimization technique which brings an update to the core artificial neuron of neural networks. Based on the modern neuroscience understanding of how biological dendrites work, this new method empowe… --- ## [AI/ML Expert for Fashion-Tech Platform](https://discuss.pytorch.org/t/ai-ml-expert-for-fashion-tech-platform/210063)
**Author:** [@leonard\_volossov](https://discuss.pytorch.org/u/leonard_volossov)\ **Replies:** 1\ **Last updated:** [March 11, 2025, 12:22pm UTC](https://discuss.pytorch.org/t/ai-ml-expert-for-fashion-tech-platform/210063 "2025-03-11T12:22:09Z")
Project Overview: We are developing a cutting-edge platform that leverages AI and ML technologies to create interactive and personalized digital experiences. Our goal is to offer a seamless and engaging experience for u… --- ## [Can't run torch conv2d with AMD Instinct MI250X](https://discuss.pytorch.org/t/cant-run-torch-conv2d-with-amd-instinct-mi250x/214120)
**Author:** [@lbj\_obj](https://discuss.pytorch.org/u/lbj_obj)\ **Replies:** 1\ **Last updated:** [March 3, 2025, 8:49pm UTC](https://discuss.pytorch.org/t/cant-run-torch-conv2d-with-amd-instinct-mi250x/214120 "2025-03-03T20:49:47Z")
I am on AMD 250X GPU after installed rocm5.6 with pytorch version 2.2 following commend pip install torch==2.2.0 torchvision==0.17.0 torchaudio==2.2.0 --index-url https://download.pytorch.org/whl/rocm5.6 I am not able… --- ## [Back propagation of a time series with LSTM](https://discuss.pytorch.org/t/back-propagation-of-a-time-series-with-lstm/217299)
**Author:** [@davide](https://discuss.pytorch.org/u/davide)\ **Replies:** 0\ **Last updated:** [February 28, 2025, 4:02pm UTC](https://discuss.pytorch.org/t/back-propagation-of-a-time-series-with-lstm/217299 "2025-02-28T16:02:32Z")
Hello everyone. I come to ask for help with a project I’m doing. I’m trying to train a neural network that needs to minimize energy consumption and I decided to start in small blocks, right now I want it to keep the tem… --- ## [Where to Post for Tensor Subclass Support in SKA?](https://discuss.pytorch.org/t/where-to-post-for-tensor-subclass-support-in-ska/217295)
**Author:** [@BouarfaMahi](https://discuss.pytorch.org/u/BouarfaMahi)\ **Replies:** 0\ **Last updated:** [February 28, 2025, 1:44pm UTC](https://discuss.pytorch.org/t/where-to-post-for-tensor-subclass-support-in-ska/217295 "2025-02-28T13:44:24Z")
I am developing Structured Knowledge Accumulation (SKA), a new framework for entropy-driven learning without backpropagation. It requires a custom tensor subclass in PyTorch to handle knowledge accumulation, decision pro… --- ## [Error:" CUDA driver initialization failed, you might not have a CUDA gpu." on an HPC](https://discuss.pytorch.org/t/error-cuda-driver-initialization-failed-you-might-not-have-a-cuda-gpu-on-an-hpc/216569)
**Author:** [@Sam\_anton](https://discuss.pytorch.org/u/Sam_anton)\ **Replies:** 1\ **Last updated:** [February 12, 2025, 1:15pm UTC](https://discuss.pytorch.org/t/error-cuda-driver-initialization-failed-you-might-not-have-a-cuda-gpu-on-an-hpc/216569 "2025-02-12T13:15:14Z")
I was trying to run rstar main.py for testing purpose on an HPC and i got this error i tried some debugging step suggested by chatgpt (rstar) \[ingenx@rdgpu01 ~\]$ nvidia-smi Wed Feb 12 13:35:14 2025 +-----------------… --- ## [Matrix factorisation using gradient descent: performing singular value decomposition on a toy dataset](https://discuss.pytorch.org/t/matrix-factorisation-using-gradient-descent-performing-singular-value-decomposition-on-a-toy-dataset/214535)
**Author:** [@akinolawilson](https://discuss.pytorch.org/u/akinolawilson)\ **Replies:** 1\ **Last updated:** [December 25, 2024, 4:22am UTC](https://discuss.pytorch.org/t/matrix-factorisation-using-gradient-descent-performing-singular-value-decomposition-on-a-toy-dataset/214535 "2024-12-25T04:22:07Z")
I’m trying to implement the singular value decomposition algorithm which factorises a matrix, M^(e,f), into three matrices U^(e,e), S^(e,f) and V^(f,f), such that M= U S V^T where U and V are orthonormal matrices and S a… --- ## [RuntimeError: The size of tensor a (307) must match the size of tensor b (12) at non-singleton dimension 2](https://discuss.pytorch.org/t/runtimeerror-the-size-of-tensor-a-307-must-match-the-size-of-tensor-b-12-at-non-singleton-dimension-2/214525)
**Author:** [@Anas\_Ayed](https://discuss.pytorch.org/u/Anas_Ayed)\ **Replies:** 1\ **Last updated:** [December 23, 2024, 1:04am UTC](https://discuss.pytorch.org/t/runtimeerror-the-size-of-tensor-a-307-must-match-the-size-of-tensor-b-12-at-non-singleton-dimension-2/214525 "2024-12-23T01:04:34Z")
i have been working on changing how DSTGN(Dynamic spatio-temporal graph network with adaptive propagation mechanism) model handles temporal dependencies the original model uses dilated temporal convolution, i tried to ch… --- ## [3dcnn neeed 4 value for stride?](https://discuss.pytorch.org/t/3dcnn-neeed-4-value-for-stride/214554)
**Author:** [@m\_rf](https://discuss.pytorch.org/u/m_rf)\ **Replies:** 2\ **Last updated:** [December 23, 2024, 1:01am UTC](https://discuss.pytorch.org/t/3dcnn-neeed-4-value-for-stride/214554 "2024-12-23T01:01:19Z")
there is some error when i use pyro and pytorch: RuntimeError: Cell In\[9\], line 54 51 # 直接使用self.conv.stride，保持3维stride 52 stride = self.conv.stride if isinstance(self.conv.stride, tuple) else (self.c… --- ## [TorchLens: New package for feature extraction and visualization of arbitrary models](https://discuss.pytorch.org/t/torchlens-new-package-for-feature-extraction-and-visualization-of-arbitrary-models/184935)
**Author:** [@johnmarktaylor](https://discuss.pytorch.org/u/johnmarktaylor)\ **Replies:** 2\ **Last updated:** [December 5, 2024, 2:54pm UTC](https://discuss.pytorch.org/t/torchlens-new-package-for-feature-extraction-and-visualization-of-arbitrary-models/184935 "2024-12-05T14:54:09Z")
I recently released a new package, TorchLens, for extracting hidden layer activations and visualizing arbitrary PyTorch models. Here’s the bullet point pitch for anyone interested: – It’s designed to work for any arbitr… --- ## [AnyModal – A Framework for Multimodal LLMs](https://discuss.pytorch.org/t/anymodal-a-framework-for-multimodal-llms/213081)
**Author:** [@ritabratamaiti](https://discuss.pytorch.org/u/ritabratamaiti)\ **Replies:** 0\ **Last updated:** [November 17, 2024, 6:21pm UTC](https://discuss.pytorch.org/t/anymodal-a-framework-for-multimodal-llms/213081 "2024-11-17T18:21:02Z")
Hi everyone, I’d like to share a project I’ve been working on: AnyModal, a modular and extensible framework for integrating diverse input modalities (like images, audio, and structured data) into large language models (… --- ## [AWS and GPU acceleration](https://discuss.pytorch.org/t/aws-and-gpu-acceleration/213057)
**Author:** [@echeyde](https://discuss.pytorch.org/u/echeyde)\ **Replies:** 0\ **Last updated:** [November 17, 2024, 2:16am UTC](https://discuss.pytorch.org/t/aws-and-gpu-acceleration/213057 "2024-11-17T02:16:32Z")
Hi, I would like to improve my inference latency by having pytorch accelerated on AWS GPUs if possible. Any recommendations as to what type of instances can be used, that is, if available/supported? Cost effectiveness is… --- ## [How I built my deep learning workstation](https://discuss.pytorch.org/t/how-i-built-my-deep-learning-workstation/118947)
**Author:** [@Roman\_Glushko](https://discuss.pytorch.org/u/Roman_Glushko)\ **Replies:** 1\ **Last updated:** [November 11, 2024, 10:06am UTC](https://discuss.pytorch.org/t/how-i-built-my-deep-learning-workstation/118947 "2024-11-11T10:06:15Z")
Recently I have managed to build my little deep learning workstation :tada: And I wanted to share my experience with you. So I wrote a blog post where tried to cover all aspects of building such a project including: r… [Previous page](https://discuss.pytorch.org/c/projects/33.md) [Next page](https://discuss.pytorch.org/c/projects/33.md?page=2)