# \#curious

**URL:** https://community.ultralytics.com/tag/curious/74.md

[Latest](https://community.ultralytics.com/latest.md) · [Categories](https://community.ultralytics.com/categories.md) · [Tags](https://community.ultralytics.com/tags.md)

---

## [Hand skeleton tracking accuracy](https://community.ultralytics.com/t/hand-skeleton-tracking-accuracy/2011)

<div class="topic-metadata">

**Author:** [@Elad\_Project](https://community.ultralytics.com/u/Elad_Project)\
**Replies:** 2\
**Last updated:** [May 31, 2026, 4:32am UTC](https://community.ultralytics.com/t/hand-skeleton-tracking-accuracy/2011 "2026-05-31T04:32:54Z")

</div>

I’m trying to train yolo11n-pose to detect hand skeletons. I’m using 900 manually annotated images (640x480) of hand joints. I’m getting a \> 0.98 pose score on validation. When looking at the resulting skeleton, it se…

---

## [YOLOv26 Beginner Questions: Human + Hand + Action Keypoints Annotation](https://community.ultralytics.com/t/yolov26-beginner-questions-human-hand-action-keypoints-annotation/1862)

<div class="topic-metadata">

**Author:** [@Time](https://community.ultralytics.com/u/Time)\
**Replies:** 1\
**Last updated:** [March 13, 2026, 8:24am UTC](https://community.ultralytics.com/t/yolov26-beginner-questions-human-hand-action-keypoints-annotation/1862 "2026-03-13T08:24:11Z")

</div>

Hi everyone, I’m a beginner with YOLOv26 and just started exploring human and hand keypoint annotation. I have several questions regarding labeling and configuration and would appreciate any guidance from the community: …

---

## [Question about PGI auxiliary branch in YOLO9 and why it wasn’t continued](https://community.ultralytics.com/t/question-about-pgi-auxiliary-branch-in-yolo9-and-why-it-wasn-t-continued/1791)

<div class="topic-metadata">

**Author:** [@djikstra0501](https://community.ultralytics.com/u/djikstra0501)\
**Replies:** 1\
**Last updated:** [February 5, 2026, 1:24pm UTC](https://community.ultralytics.com/t/question-about-pgi-auxiliary-branch-in-yolo9-and-why-it-wasn-t-continued/1791 "2026-02-05T13:24:53Z")

</div>

YOLO9 introduced the PGI (Programmable Gradient Information) auxiliary branch during training, which seemed like a principled way to improve gradient flow and representation learning, especially for deeper architectures. …

---

## [Does Ultralytics support model Pruning?](https://community.ultralytics.com/t/does-ultralytics-support-model-pruning/1680)

<div class="topic-metadata">

**Author:** [@AA278-HUB](https://community.ultralytics.com/u/AA278-HUB)\
**Replies:** 3\
**Last updated:** [December 3, 2025, 1:30pm UTC](https://community.ultralytics.com/t/does-ultralytics-support-model-pruning/1680 "2025-12-03T13:30:45Z")

</div>

and After pruning my YOLO model, Ultralytics can’t load it due to shape/key mismatches. Can pruned models still be validated with model.val()?

---

## [Can YOLO support incremental training to add new data or classes without starting from scratch?](https://community.ultralytics.com/t/can-yolo-support-incremental-training-to-add-new-data-or-classes-without-starting-from-scratch/1684)

<div class="topic-metadata">

**Author:** [@kirti](https://community.ultralytics.com/u/kirti)\
**Replies:** 1\
**Last updated:** [December 3, 2025, 7:04am UTC](https://community.ultralytics.com/t/can-yolo-support-incremental-training-to-add-new-data-or-classes-without-starting-from-scratch/1684 "2025-12-03T07:04:03Z")

</div>

Hi, I’m working with YOLO models and want to know whether they support incremental training for new classes. Is it possible to add additional object classes to an already trained model without retraining from scratch o…

---

## [Request for YOLOv12 Architecture & Training Data Details for Research Reproduction](https://community.ultralytics.com/t/request-for-yolov12-architecture-training-data-details-for-research-reproduction/1668)

<div class="topic-metadata">

**Author:** [@Abdul\_Nafay](https://community.ultralytics.com/u/Abdul_Nafay)\
**Replies:** 1\
**Last updated:** [November 27, 2025, 5:51am UTC](https://community.ultralytics.com/t/request-for-yolov12-architecture-training-data-details-for-research-reproduction/1668 "2025-11-27T05:51:33Z")

</div>

Hello Ultralytics Team, I am conducting a research study where I aim to: Reproduce the YOLOv12 training process, Analyze architectural variations, and Evaluate how modifications affect performance and generaliz…

---

## [Can I use SAHI Tiled Inference with YOLO11 Segmentation Model?](https://community.ultralytics.com/t/can-i-use-sahi-tiled-inference-with-yolo11-segmentation-model/1559)

<div class="topic-metadata">

**Author:** [@autkucakan](https://community.ultralytics.com/u/autkucakan)\
**Replies:** 1\
**Last updated:** [October 20, 2025, 4:03pm UTC](https://community.ultralytics.com/t/can-i-use-sahi-tiled-inference-with-yolo11-segmentation-model/1559 "2025-10-20T16:03:25Z")

</div>

I have a segmentation dataset for my project and I thought if I could use SAHI with the segmentation model because it’s looking beneficial to me but there is no source I could figure out that from. In docs I could only f…

---

## [Low-Level Details of Post-Training INT8 Quantization and Calibration](https://community.ultralytics.com/t/low-level-details-of-post-training-int8-quantization-and-calibration/1506)

<div class="topic-metadata">

**Author:** [@Kallinteris-Andreas](https://community.ultralytics.com/u/Kallinteris-Andreas)\
**Replies:** 1\
**Last updated:** [September 26, 2025, 1:58pm UTC](https://community.ultralytics.com/t/low-level-details-of-post-training-int8-quantization-and-calibration/1506 "2025-09-26T13:58:59Z")

</div>

I’m looking for a low-level explanation of how post-training quantization (PTQ) and, specifically, the INT8 calibration process works. My goal is to understand the underlying mechanics and best practices. Here are my sp…

---

## [Estimating Computational Power for 75 AI-Powered Cameras](https://community.ultralytics.com/t/estimating-computational-power-for-75-ai-powered-cameras/647)

<div class="topic-metadata">

**Author:** [@Fahim\_Anwar](https://community.ultralytics.com/u/Fahim_Anwar)\
**Replies:** 5\
**Last updated:** [May 12, 2025, 11:30am UTC](https://community.ultralytics.com/t/estimating-computational-power-for-75-ai-powered-cameras/647 "2025-05-12T11:30:57Z")

</div>

Hi Ultralytics Community, I hope you’re all doing well! I am currently working on a project where the client aims to deploy 75 AI-powered cameras. These cameras will perform two tasks: Detecting intrusions inside a v…

---

## [Output of the model in training vs. inference](https://community.ultralytics.com/t/output-of-the-model-in-training-vs-inference/172)

<div class="topic-metadata">

**Author:** [@robin\_rob96](https://community.ultralytics.com/u/robin_rob96)\
**Replies:** 1\
**Last updated:** [August 7, 2024, 1:34pm UTC](https://community.ultralytics.com/t/output-of-the-model-in-training-vs-inference/172 "2024-08-07T13:34:46Z")

</div>

Hi, I just read the yolo paper from 2015 in which it states that the predictions are encoded in a tensor of shape SxSx(5\*B + C) So that means that for every cell in the grid there are 5 params per box + probabilities f…

---

## [Set validation set the same as training set](https://community.ultralytics.com/t/set-validation-set-the-same-as-training-set/83)

<div class="topic-metadata">

**Author:** [@Nima\_Chelongar](https://community.ultralytics.com/u/Nima_Chelongar)\
**Replies:** 8\
**Last updated:** [July 16, 2024, 8:24am UTC](https://community.ultralytics.com/t/set-validation-set-the-same-as-training-set/83 "2024-07-16T08:24:11Z")

</div>

hi guys. i have trained model with YOLOV8 in which used training set as my validation set, but i get very good result over unseen data. now when i split the data into the training and validating, i get overfitted model…

---

## [What do you use to write code?](https://community.ultralytics.com/t/what-do-you-use-to-write-code/89)

<div class="topic-metadata">

**Author:** [@BurhanQ](https://community.ultralytics.com/u/BurhanQ)\
**Replies:** 1\
**Last updated:** [July 11, 2024, 8:17pm UTC](https://community.ultralytics.com/t/what-do-you-use-to-write-code/89 "2024-07-11T20:17:10Z")

</div>

poll
