Hi all, Description I want to train detectnet-v2 + res 18 on kitti dataset, I mounted the docker container on host system and load the kitti dataset as offline and then convert to tfrecord and then train it only one epoch. and I followed from TLT document. Matching predictions to ground truth, cla…

I also trigger the same training. Totally run 1 epoch. I can see the loss will reduce gracefully. It is correct. 0.12829292 → 0.11809954 → 0.06464981 → 0.011154257 → 0.0026569464 → 0.0019884342 → … See https://docs.nvidia.com/metropolis/TLT/tlt-getting-started-guide/index.html#model This par…

1- You trained DetectNet-v2+resnet18 on kitti with size 1248*348? 2- You resized all images before creating tfrecord? 3- I need to set flag load_graph to True if I define the path of pretrained model?

Please refer to the default jupyter notebook inside the docker. should be 1248* 384 No needed to resize because the KITTI dataset are mostly 1248*384 For training, it is not needed. For retraining, it is needed. See details in tlt user guide.

@Morganh , I seem to be training a custom dataset and I face the exact same problem. Any idea

Train detectnet-v2 resnet 18 on kitti dataset

Accelerated Computing Intelligent Video Analytics TAO Toolkit

Morganh August 27, 2020, 5:53am 6

@beefshepherd
For your question, let us track with your created topic Tlt detectnet training focusing on a particular class? - #5 by beefshepherd

Topic		Replies	Views
TLT-Train DetectNetv2 ResNet18 always give mAP 0% at target class TAO Toolkit	11	663	October 26, 2022
Training with TLT a detectnet_v2 resnet18 pre-trained model failed TAO Toolkit	2	678	October 12, 2021
Training Custom Object detector with 6 classes TAO Toolkit	27	2522	October 12, 2021
Training Detectnet_v2 on darknet annotation format dataset TAO Toolkit ai-training	3	773	October 12, 2021
Object Detection using TAO DetectNet_v2. The category accuracy results are missing TAO Toolkit	17	891	January 13, 2022
Retrained model shows worse results compared to NVIDIA provided model TAO Toolkit	16	450	May 29, 2024
tlt-train error when deploy mobilenet_v2 by using DetectNet TAO Toolkit	28	2797	October 12, 2021
TLT - retrain trafficcamnet with customized data precision is 0 TAO Toolkit	20	907	October 12, 2021
Errors in Training, 0 or Nan mAP, Low Loss, Tutorial Config TAO Toolkit	35	2141	October 12, 2021
TLT for custom dataset and deploy .engine file TAO Toolkit	2	1010	October 12, 2021

Train detectnet-v2 resnet 18 on kitti dataset

Related topics