# Show and Tell

**URL:** https://discuss.ai.google.dev/c/tensorflow/show-tell/30.md

[Latest](https://discuss.ai.google.dev/latest.md) · [Categories](https://discuss.ai.google.dev/categories.md) · [Tags](https://discuss.ai.google.dev/tags.md)

---

## [Official #MadeWithTFJS Show & Tell inspiration thread](https://discuss.ai.google.dev/t/official-madewithtfjs-show-tell-inspiration-thread/30761)

<div class="topic-metadata">

**Author:** [@Jason](https://discuss.ai.google.dev/u/Jason)\
**Replies:** 22\
**Last updated:** [January 27, 2025, 6:07pm UTC](https://discuss.ai.google.dev/t/official-madewithtfjs-show-tell-inspiration-thread/30761 "2025-01-27T18:07:42Z")

</div>

Once per quarter we meet with several members from the TensorFlow.js community to learn more about what amazing projects they have created. This thread is the one stop place to see the latest interviews to get inspiratio…

---

## [About the Show and Tell category](https://discuss.ai.google.dev/t/about-the-show-and-tell-category/30504)

<div class="topic-metadata">

**Author:** [@carlos](https://discuss.ai.google.dev/u/carlos)\
**Replies:** 2\
**Last updated:** [June 17, 2024, 5:32am UTC](https://discuss.ai.google.dev/t/about-the-show-and-tell-category/30504 "2024-06-17T05:32:42Z")

</div>

Got a cool project? Share and show off your work. Feel free to share links to demos, GitHub and posts for your project. You can also submit your project to the TensorFlow Community Spotlight program for the chance to be…

---

## [ONNX Probabilistic Programming Working Group – Collaboration with the TensorFlow Probability Community](https://discuss.ai.google.dev/t/onnx-probabilistic-programming-working-group-collaboration-with-the-tensorflow-probability-community/129528)

<div class="topic-metadata">

**Author:** [@Brian\_Parbhu](https://discuss.ai.google.dev/u/Brian_Parbhu)\
**Replies:** 0\
**Last updated:** [March 7, 2026, 8:19pm UTC](https://discuss.ai.google.dev/t/onnx-probabilistic-programming-working-group-collaboration-with-the-tensorflow-probability-community/129528 "2026-03-07T20:19:37Z")

</div>

Hi everyone, I wanted to share that the ONNX Probabilistic Programming Working Group has recently been formed and invite participation from the TensorFlow Probability (TFP) community. The goal of this working group is …

---

## [Happy new year to all of us](https://discuss.ai.google.dev/t/happy-new-year-to-all-of-us/118294)

<div class="topic-metadata">

**Author:** [@patrickbiyaga](https://discuss.ai.google.dev/u/patrickbiyaga)\
**Replies:** 8\
**Last updated:** [January 25, 2026, 7:58am UTC](https://discuss.ai.google.dev/t/happy-new-year-to-all-of-us/118294 "2026-01-25T07:58:02Z")

</div>

Happy new year to all. This year will mark my total immersion in the field of artificial intelligence. I’d be delighted to share my impressions of my reading and practical experiences here.

---

## [GRA+LLM Hybrid Architecture — Technical Overview, Formulas, Advantages, and Simulator Links](https://discuss.ai.google.dev/t/gra-llm-hybrid-architecture-technical-overview-formulas-advantages-and-simulator-links/109140)

<div class="topic-metadata">

**Author:** [@AAA\_AAA](https://discuss.ai.google.dev/u/AAA_AAA)\
**Replies:** 0\
**Last updated:** [November 9, 2025, 2:21pm UTC](https://discuss.ai.google.dev/t/gra-llm-hybrid-architecture-technical-overview-formulas-advantages-and-simulator-links/109140 "2025-11-09T14:21:11Z")

</div>

Forum Topic Update: GRA+LLM Hybrid Architecture — Technical Overview, Formulas, Advantages, and Simulator Links Introduction The GRA+LLM Hybrid Architecture fuses explicit graph-based reasoning (GRA) with neural gener…

---

## [I wrote some optimizers for TensorFlow](https://discuss.ai.google.dev/t/i-wrote-some-optimizers-for-tensorflow/107418)

<div class="topic-metadata">

**Author:** [@NoteDance](https://discuss.ai.google.dev/u/NoteDance)\
**Replies:** 0\
**Last updated:** [October 13, 2025, 10:16am UTC](https://discuss.ai.google.dev/t/i-wrote-some-optimizers-for-tensorflow/107418 "2025-10-13T10:16:22Z")

</div>

Hello everyone, I wrote some optimizers for TensorFlow. If you’re using TensorFlow, they should be helpful to you. https://github.com/NoteDance/optimizers

---

## [Time series saliency maps with Cross-Domain Integrated Gradients](https://discuss.ai.google.dev/t/time-series-saliency-maps-with-cross-domain-integrated-gradients/103791)

<div class="topic-metadata">

**Author:** [@kechrisc](https://discuss.ai.google.dev/u/kechrisc)\
**Replies:** 1\
**Last updated:** [September 16, 2025, 9:27am UTC](https://discuss.ai.google.dev/t/time-series-saliency-maps-with-cross-domain-integrated-gradients/103791 "2025-09-16T09:27:10Z")

</div>

We are introducing a saliency map method that produces meaningful attributions for time-series ML models. Instead of restricting explanations to the raw time domain, our approach generates saliency maps in relevant domai…

---

## [A lightweight utility for training multiple Keras models in parallel and comparing their final loss and last-epoch time](https://discuss.ai.google.dev/t/a-lightweight-utility-for-training-multiple-keras-models-in-parallel-and-comparing-their-final-loss-and-last-epoch-time/87275)

<div class="topic-metadata">

**Author:** [@NoteDance](https://discuss.ai.google.dev/u/NoteDance)\
**Replies:** 0\
**Last updated:** [June 5, 2025, 10:28am UTC](https://discuss.ai.google.dev/t/a-lightweight-utility-for-training-multiple-keras-models-in-parallel-and-comparing-their-final-loss-and-last-epoch-time/87275 "2025-06-05T10:28:45Z")

</div>

---

## [This Python class offers a multiprocessing-powered Pool for efficiently collecting and managing experience replay data in reinforcement learning](https://discuss.ai.google.dev/t/this-python-class-offers-a-multiprocessing-powered-pool-for-efficiently-collecting-and-managing-experience-replay-data-in-reinforcement-learning/86672)

<div class="topic-metadata">

**Author:** [@NoteDance](https://discuss.ai.google.dev/u/NoteDance)\
**Replies:** 0\
**Last updated:** [June 2, 2025, 12:28pm UTC](https://discuss.ai.google.dev/t/this-python-class-offers-a-multiprocessing-powered-pool-for-efficiently-collecting-and-managing-experience-replay-data-in-reinforcement-learning/86672 "2025-06-02T12:28:40Z")

</div>

---

## [Weka vs Tensorflow](https://discuss.ai.google.dev/t/weka-vs-tensorflow/84147)

<div class="topic-metadata">

**Author:** [@Carles\_Vila](https://discuss.ai.google.dev/u/Carles_Vila)\
**Replies:** 0\
**Last updated:** [May 21, 2025, 10:08am UTC](https://discuss.ai.google.dev/t/weka-vs-tensorflow/84147 "2025-05-21T10:08:06Z")

</div>

Hello, Here I wanted to discuss about some much better results that I got with Weka compared to Tensorflow. ¿Would be it possible to replicate exactly the Weka MLP model I used in Tensorflow? I wanted to discriminate b…

---

## [: A Diagnostic SDK to Fix TensorFlow Memory Residues (Eclipse Leaks, Orphan Threads, CUDA Artifacts](https://discuss.ai.google.dev/t/a-diagnostic-sdk-to-fix-tensorflow-memory-residues-eclipse-leaks-orphan-threads-cuda-artifacts/83216)

<div class="topic-metadata">

**Author:** [@Hussein\_Shtia](https://discuss.ai.google.dev/u/Hussein_Shtia)\
**Replies:** 0\
**Last updated:** [May 14, 2025, 7:43am UTC](https://discuss.ai.google.dev/t/a-diagnostic-sdk-to-fix-tensorflow-memory-residues-eclipse-leaks-orphan-threads-cuda-artifacts/83216 "2025-05-14T07:43:57Z")

</div>

Dear TensorFlow Community, We’ve developed and released a diagnostic SDK called CollapseCleaner to address a class of runtime issues commonly reported but hard to reproduce — namely: Retained background threads (e.g.,…

---

## [TensorFlow implementation for optimizers](https://discuss.ai.google.dev/t/tensorflow-implementation-for-optimizers/81407)

<div class="topic-metadata">

**Author:** [@NoteDance](https://discuss.ai.google.dev/u/NoteDance)\
**Replies:** 0\
**Last updated:** [April 29, 2025, 2:53pm UTC](https://discuss.ai.google.dev/t/tensorflow-implementation-for-optimizers/81407 "2025-04-29T14:53:40Z")

</div>

Hello everyone, I implement some optimizers using TensorFlow. I hope this project can help you.

---

## [How to extract multi-channel features of independent variables to predict the value distribution of a single channel dependent variable](https://discuss.ai.google.dev/t/how-to-extract-multi-channel-features-of-independent-variables-to-predict-the-value-distribution-of-a-single-channel-dependent-variable/80577)

<div class="topic-metadata">

**Author:** [@Linchen](https://discuss.ai.google.dev/u/Linchen)\
**Replies:** 0\
**Last updated:** [April 22, 2025, 9:11am UTC](https://discuss.ai.google.dev/t/how-to-extract-multi-channel-features-of-independent-variables-to-predict-the-value-distribution-of-a-single-channel-dependent-variable/80577 "2025-04-22T09:11:24Z")

</div>

Machine Learning Training Statement Purpose of Machine Learning Training The purpose of the machine learning training is to study the impact of urban road distribution, population density distribution, and the distributi…

---

## [MONAI-like library in Keras for medical image processing](https://discuss.ai.google.dev/t/monai-like-library-in-keras-for-medical-image-processing/76807)

<div class="topic-metadata">

**Author:** [@innat](https://discuss.ai.google.dev/u/innat)\
**Replies:** 0\
**Last updated:** [April 1, 2025, 11:24am UTC](https://discuss.ai.google.dev/t/monai-like-library-in-keras-for-medical-image-processing/76807 "2025-04-01T11:24:10Z")

</div>

I’ve released a Keras 3-based AI library for medical image processing. Inspired by ‘MONAI’, it is designed to streamline deep learning workflows for medical imaging, offering pre-built models, data augmentation technique…

---

## [68.2X faster than TensorFlow](https://discuss.ai.google.dev/t/68-2x-faster-than-tensorflow/64180)

<div class="topic-metadata">

**Author:** [@Amenti\_Group](https://discuss.ai.google.dev/u/Amenti_Group)\
**Replies:** 0\
**Last updated:** [February 3, 2025, 8:55am UTC](https://discuss.ai.google.dev/t/68-2x-faster-than-tensorflow/64180 "2025-02-03T08:55:34Z")

</div>

Someone shipped a code that outperforms TensorFlow by significant numbers! Can someone validate this! Its a big breakthrough if this is true. Its open-sources! Github: GitHub - tsotchke/quantum\_geometric\_tensor: High …

---

## [Updates on Train a Deep Q Network with TF-Agents](https://discuss.ai.google.dev/t/updates-on-train-a-deep-q-network-with-tf-agents/61284)

<div class="topic-metadata">

**Author:** [@Hernando\_De\_Avila\_Pe](https://discuss.ai.google.dev/u/Hernando_De_Avila_Pe)\
**Replies:** 0\
**Last updated:** [January 17, 2025, 4:26pm UTC](https://discuss.ai.google.dev/t/updates-on-train-a-deep-q-network-with-tf-agents/61284 "2025-01-17T16:26:03Z")

</div>

Hi, I have been reading the tutorial Train a Deep Q Network with TF-Agents Fine by now, but I run into a problem, look like the resent update of TensorFlow are not compatible with reverb version in !pip install tf-agent…

---

## [I wrote optimizers for TensorFlow and Keras](https://discuss.ai.google.dev/t/i-wrote-optimizers-for-tensorflow-and-keras/58898)

<div class="topic-metadata">

**Author:** [@NoteDance](https://discuss.ai.google.dev/u/NoteDance)\
**Replies:** 0\
**Last updated:** [January 4, 2025, 11:30am UTC](https://discuss.ai.google.dev/t/i-wrote-optimizers-for-tensorflow-and-keras/58898 "2025-01-04T11:30:36Z")

</div>

Hello everyone, I wrote optimizers for TensorFlow and Keras, and they are used in the same way as Keras optimizers.

---

## [Left ventricular function classification of apical 4-chamber view echocardiogram videos](https://discuss.ai.google.dev/t/left-ventricular-function-classification-of-apical-4-chamber-view-echocardiogram-videos/57178)

<div class="topic-metadata">

**Author:** [@Nabeel\_Amiruddin](https://discuss.ai.google.dev/u/Nabeel_Amiruddin)\
**Replies:** 0\
**Last updated:** [December 25, 2024, 1:16am UTC](https://discuss.ai.google.dev/t/left-ventricular-function-classification-of-apical-4-chamber-view-echocardiogram-videos/57178 "2024-12-25T01:16:14Z")

</div>

I have pre-processed echocardiograms videos into arrays with shape (128,112,112,1) each and am trying to classify the left ventricular function with supervised learning with a 3D convolutional network. A simple 2 conv3D …

---

## [Image classification with JAX & Flax](https://discuss.ai.google.dev/t/image-classification-with-jax-flax/30659)

<div class="topic-metadata">

**Author:** [@MWITIDERRICK](https://discuss.ai.google.dev/u/MWITIDERRICK)\
**Replies:** 0\
**Last updated:** [July 8, 2022, 11:37am UTC](https://discuss.ai.google.dev/t/image-classification-with-jax-flax/30659 "2022-07-08T11:37:11Z")

</div>

Learn how to build convolutional neural networks with JAX and Flax. Flax is a neural network library for JAX. JAX is a Python library that provides high-performance computing in machine learning research. JAX provides an…

---

## [Open-Source Pose Estimation Movement Analysis for mobile & web applications](https://discuss.ai.google.dev/t/open-source-pose-estimation-movement-analysis-for-mobile-web-applications/53814)

<div class="topic-metadata">

**Author:** [@Fabrice\_Sepret](https://discuss.ai.google.dev/u/Fabrice_Sepret)\
**Replies:** 0\
**Last updated:** [December 11, 2024, 8:41am UTC](https://discuss.ai.google.dev/t/open-source-pose-estimation-movement-analysis-for-mobile-web-applications/53814 "2024-12-11T08:41:56Z")

</div>

Hello TensorFlow Community, I’m reaching out to discuss an exciting potential open-source project in pose estimation and movement analysis. Project Context With PoseTracker API, we’ve developed a streamlined approach b…

---

## [Convert Any PyTorch ML Model to TensorFlow, JAX, or NumPy with Ivy! 🚀 + New Kornia Integration](https://discuss.ai.google.dev/t/convert-any-pytorch-ml-model-to-tensorflow-jax-or-numpy-with-ivy-new-kornia-integration/48163)

<div class="topic-metadata">

**Author:** [@Yusha\_Arif](https://discuss.ai.google.dev/u/Yusha_Arif)\
**Replies:** 0\
**Last updated:** [November 7, 2024, 4:35pm UTC](https://discuss.ai.google.dev/t/convert-any-pytorch-ml-model-to-tensorflow-jax-or-numpy-with-ivy-new-kornia-integration/48163 "2024-11-07T16:35:44Z")

</div>

Hey everyone! Just wanted to share something exciting for those of you working across multiple ML frameworks. Ivy is a Python package that allows you to seamlessly convert ML models and code between frameworks like PyTo…

---

## [Next in the tutorial - Sentiment Analysis](https://discuss.ai.google.dev/t/next-in-the-tutorial-sentiment-analysis/37135)

<div class="topic-metadata">

**Author:** [@Kelly\_McTiernan](https://discuss.ai.google.dev/u/Kelly_McTiernan)\
**Replies:** 2\
**Last updated:** [August 24, 2024, 9:34pm UTC](https://discuss.ai.google.dev/t/next-in-the-tutorial-sentiment-analysis/37135 "2024-08-24T21:34:49Z")

</div>

The next example was the Movie Review Sentiment Analysis. It did terribly with test data, in particular with 5 unseen, short reviews. I modified it by adding a bidirectional LSTM layer, added early stopping, and annoyi…

---

## [A machine learning library that allows you to easily train agents](https://discuss.ai.google.dev/t/a-machine-learning-library-that-allows-you-to-easily-train-agents/35609)

<div class="topic-metadata">

**Author:** [@NoteDance](https://discuss.ai.google.dev/u/NoteDance)\
**Replies:** 0\
**Last updated:** [August 8, 2024, 5:43am UTC](https://discuss.ai.google.dev/t/a-machine-learning-library-that-allows-you-to-easily-train-agents/35609 "2024-08-08T05:43:34Z")

</div>

Hello, everyone, this is my project, it allows you to easily train agents.

---

## [A simplistic flax trainer](https://discuss.ai.google.dev/t/a-simplistic-flax-trainer/27567)

<div class="topic-metadata">

**Author:** [@Nithish\_M](https://discuss.ai.google.dev/u/Nithish_M)\
**Replies:** 0\
**Last updated:** [June 24, 2024, 5:46am UTC](https://discuss.ai.google.dev/t/a-simplistic-flax-trainer/27567 "2024-06-24T05:46:08Z")

</div>

I recently started migrating from Tensorflow to Jax , Flax. I’ve created a new repository to help people transition from TensorFlow to JAX and Flax. It features a simple Flax trainer that’s perfect for beginners wanting…

---

## [Java Object Detection Helper](https://discuss.ai.google.dev/t/java-object-detection-helper/31646)

<div class="topic-metadata">

**Author:** [@preafixed](https://discuss.ai.google.dev/u/preafixed)\
**Replies:** 9\
**Last updated:** [June 11, 2024, 2:21pm UTC](https://discuss.ai.google.dev/t/java-object-detection-helper/31646 "2024-06-11T14:21:03Z")

</div>

Hey there, I build a library for using TensorFlows shared models in Java, and running inference on them. I saw a lot of people struggling with inference in Java, so I created Java-TF. Manage and load different saved m…

---

## [I wrote a machine learning library based on TensorFlow](https://discuss.ai.google.dev/t/i-wrote-a-machine-learning-library-based-on-tensorflow/30565)

<div class="topic-metadata">

**Author:** [@NoteDancing](https://discuss.ai.google.dev/u/NoteDancing)\
**Replies:** 0\
**Last updated:** [June 4, 2024, 11:45am UTC](https://discuss.ai.google.dev/t/i-wrote-a-machine-learning-library-based-on-tensorflow/30565 "2024-06-04T11:45:48Z")

</div>

Note is a system(library) for deep learning and reinforcement learning. Note makes the building and training of neural networks easy and flexible. Note.nn.layer package contains many layer modules, you can use them to bu…

---

## [A 14-hour beginner-friendly introduction to TensorFlow](https://discuss.ai.google.dev/t/a-14-hour-beginner-friendly-introduction-to-tensorflow/31933)

<div class="topic-metadata">

**Author:** [@mrdbourke](https://discuss.ai.google.dev/u/mrdbourke)\
**Replies:** 0\
**Last updated:** [May 30, 2021, 9:56pm UTC](https://discuss.ai.google.dev/t/a-14-hour-beginner-friendly-introduction-to-tensorflow/31933 "2021-05-30T21:56:16Z")

</div>

I’ve created a 14-hour tutorial introduction to deep learning fundamentals + TensorFlow. It covers: The ins and outs of Google Colab TensorFlow basics (tensor creation, tensor manipulation, tensor aggregation) Deep le…

---

## [Bird classification model (in progress)](https://discuss.ai.google.dev/t/bird-classification-model-in-progress/28922)

<div class="topic-metadata">

**Author:** [@CosmicBee](https://discuss.ai.google.dev/u/CosmicBee)\
**Replies:** 2\
**Last updated:** [May 28, 2024, 11:18pm UTC](https://discuss.ai.google.dev/t/bird-classification-model-in-progress/28922 "2024-05-28T23:18:28Z")

</div>

Hi there, I’ve been working on a bird classification model. I’m hoping to get it onto my TinyML hardware but the RAM requirements are currently too large as I used a rather large input size. I’m working to further adjus…

---

## [Monitor your ML jobs on mobile phones📱](https://discuss.ai.google.dev/t/monitor-your-ml-jobs-on-mobile-phones/29943)

<div class="topic-metadata">

**Author:** [@rishit\_dagli](https://discuss.ai.google.dev/u/rishit_dagli)\
**Replies:** 1\
**Last updated:** [May 9, 2024, 9:05am UTC](https://discuss.ai.google.dev/t/monitor-your-ml-jobs-on-mobile-phones/29943 "2024-05-09T09:05:01Z")

</div>

What if you could monitor your Machine Learning jobs on Google Colab or Kaggle or pretty much anywhere in real-time on your mobile phones📱. You can now monitor your ML jobs even while taking a walk🚶with \<5 lines of code …

---

## [Fine-tuning Stable Diffusion with KerasCV 🔥](https://discuss.ai.google.dev/t/fine-tuning-stable-diffusion-with-kerascv/29594)

<div class="topic-metadata">

**Author:** [@Sayak\_Paul](https://discuss.ai.google.dev/u/Sayak_Paul)\
**Replies:** 5\
**Last updated:** [May 6, 2024, 5:40am UTC](https://discuss.ai.google.dev/t/fine-tuning-stable-diffusion-with-kerascv/29594 "2024-05-06T05:40:24Z")

</div>

Hi folks, I hope y’all are doing well. Happy new year! Delighted to present a project that @deep-diver and I have been working on for a while now – Fine-tuning Stable Diffusion with KerasCV. The project shows how to f…

[Next page](https://discuss.ai.google.dev/c/tensorflow/show-tell/30.md?page=1)
