Fine-tuning TrafficCamNet RT-DETR: Adding 4 New Classes to a 5-Class...
Fine-tuning TrafficCamNet RT-DETR: Adding 4 New Classes to a 5-Class Pretrained Model Hello NVIDIA Team, I am working with the NVIDIA TrafficCamNet RT-DETR model and using the TAO Toolkit for...
View ArticleTrafficCamNet Transformer Lite (RT-DETR ResNet-50) very slow on RTX 4060 8GB...
Hello NVIDIA team, I am fine-tuning TrafficCamNet Transformer Lite (RT-DETR with ResNet-50 backbone) on a custom COCO-format object-detection dataset using TAO Toolkit 6.26.3. The model trains...
View ArticleMask Grounding DINO CC v2.1 - Request for version-specific lineage,...
This is not an error report or troubleshooting request. It is a version-specific artifact lineage, licensing, and runtime-evidence question submitted before acquisition or execution. - **Hardware:**...
View ArticleUnable to train YOLOv6 model with TAO
I am currently working on the NVIDIA Jetson Orin Nano Developer Kit and building a real-time safety monitoring system using YOLOv6 for person detection. I also plan to implement transfer learning to...
View ArticleSegmentation on instances with holes in an instance mask
I tried training a model using Nvidia Tao mask2former with images that have rle annotations and iscrowd=0 (instances that have holes in the mask or instances that have disconnected islands), but the...
View Article[Regression] nvmultiurisrcbin does not trigger RTSP reconnect monitoring on...
Hello, sorry to disturb you Hardware Platform: x86 dGPU GPU: Tesla T4 NVIDIA Driver Version: 590.48.01 Operating System: Ubuntu 24.04.4 LTS DeepStream Version: 9.0 and 9.1 Application: Python /...
View ArticleSegformer
What is the parameter to use class weights to balance two or more classes in nvidia tao segformer (5.5.0 and 5.3.0) specs? 1 post - 1 participant Read full topic
View ArticleTraining BodyPoseNet Model by Freezing the backbone layers of VGG Architecture
Please provide the following information when requesting support. • NVIDIA RTX 6000 Ada • BodyPoseNet Model • trainable_v1.0 bpnet_train_m1_coco.txt (2.9 KB) We want to know and confirmation from you...
View ArticleRT-DETR (TAO 6.x) INT8 accuracy drops than FP16/FP32
Environment GPU: NVIDIA RTX 2060 (6 GB, Turing, CC 7.5) Driver / CUDA: 580.x / CUDA 13.0 TAO Toolkit: 6.25.11 (6.25.11-pyt for train/export/quantize, 6.25.11-deploy for gen_trt_engine) TensorRT in TAO...
View ArticleClik here to view.
TAO Toolkit 6.0 wandb Integration: TensorBoard Dashboard Not Loading in wandb
Please provide the following information when requesting support. • Hardware : RTX A4000 • Network Type : yolo_v3 • TLT Version : nvcr.io/nvidia/tao/tao-toolkit:5.0.0-tf1.15.5 • Docker run command :...
View ArticleNVIDIA TAO 7 is now available
NVIDIA TAO 7 is now generally available, adding a suite of agent skills and tools for fine-tuning vision AI models with coding agents and natural language prompts. What’s new: Agent skills and tools...
View Article[TAO Deploy] INT8 Calibration Fails for RT-DETR with...
Dear @Morganh I trained teacher and student model inside 6.26 containers and then I created notebook for training using tao. I am able to train model both teacher and student and able to generate fp16...
View ArticleClik here to view.
Query on PeopleNet-Transformer model constantly misses bounding boxes -...
Hi Team, This is in reference to below issue, Query on Multi-Camera Calibration Accuracy vs Unified ID Reassociation Stability in MV3DT DeepStream SDK Hi Kesong, Can you please help me with TAO forum...
View ArticleIssue while run rtdetr distill command nvidia tao
Hi Team, I am using Nvidia Tao toolkit nvcr.io/nvidia/tao/tao-toolkit:6.26.3-pyt and trying to train rt-detr-distill model. I have prepared my dataset in coco format. I have use below model as a...
View ArticleClik here to view.
Tao deploy make a rabbish
Please provide the following information when requesting support. • Hardware (x86_64) • Network Type (RTDetr) Good day. I decided to run a model downloaded from the NGC catalog ( GPU-optimized AI,...
View ArticleRTDETR FP16 on Deepstream produces NaN Values
Continuing the discussion from The RTDETR model is used on DeepStream with FP16 precision and no bounding boxes: Please provide the following information when requesting support. • Hardware NVIDIA...
View ArticleClik here to view.
Tao Toolkit. Error Responce from Daemon
Hello. I’m getting an error when trying to register in docker nvcr.io. It doesn’t work on both my corporate network and mobile internet. However, it was working fine yesterday. I’m using a Legacy key...
View ArticleClik here to view.
Problems with TAO Toolkit
• Hardware GPU with WSL • Network Type - • TLT Version: - • How to reproduce the issue ? Good afternoon. I’m facing a problem when trying to run a model using the Tao launcher. When I run tao model...
View ArticleProblem with fine-tuning using gd-tiny model
We have a market project targeting object detection scenarios, where the customer provided multiple scenario samples. Currently, we are using the TAO tool with the built-in GD-TINY CV model for...
View ArticleSomething wrong with tao deploy
There is something wrong with tao deploy, maybe the docker nvcr.io/nvidia/tao/tao-toolkit:6.25.11-deploy (launcher) by@System-Product-Name:~/tao_tutorials$ tao deploy rtdetr gen_trt_engine -e...
View Article
![[-100%] Commodity Intelligence – (Android Apps)](http://play-lh.googleusercontent.com/j67-qJvmnn4xnVlYVZ9ZhtZUJIgwjKekJGoKQT6D1KHcWGZxtQaFroummEvwV17na7tyh1pGbbeWCRdWtTKqeg=w950-h950)
