TAO Toolkit 6.0 wandb Integration: TensorBoard Dashboard Not Loading in wandb

Accelerated Computing Intelligent Video Analytics TAO Toolkit
1

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 : docker run --gpus all -it --rm -e WANDB_API_KEY=<api_key> -e WANDB_PROJECT=yolo_v3 -e WANDB_TENSORBOARD=true -v /home/user/Desktop/tao/cv_samples_v1.4.0/yolo_v3/:/workspace/tao-experiments -v /home/user/Desktop/tao/cv_samples_v1.4.0/yolo_v3/specs:/workspace/tao-experiments/yolo_v3/specs nvcr.io/nvidia/tao/tao-toolkit:5.0.0-tf1.15.5 /bin/bash

I am trying to integrate Weights & Biases (W&B) with NVIDIA TAO Toolkit 6.0 by following the official TAO documentation. The W&B integration itself is working correctly: training runs, logs, and configuration data are successfully uploaded and visible in the W&B dashboard.

However, the TensorBoard integration inside W&B is not working

Screenshot from 2026-06-30 17-55-48
Screenshot from 2026-06-30 17-55-481854×859 148 KB

yolo_v3_train_resnet18_tfrecord.txt (2.0 KB)

3

Could you please try to refer to TAO WandB Integration — Tao Toolkit to add wandb section? For example,

wandb {
  enabled: true
  project: "face_detection"
  entity: "yourname"
  tags: "training"
  tags: "tao_toolkit"
  reinit: true
  sync_tensorboard: true
  save_code: false
  name: "detection"
}
4

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks.