I trained 300 images with mask rcnn but the results are very poor. The mAP is up to 40% with other models than TAO. This is my results after 140 iterations: DLL 2021-09-19 16:26:47.189391 - Iteration: 140 Validation Iteration: 140 AP : 0.017194924876093864 DLL 2021-09-19 16:26:47.189477 - Iteratio…

That reference is talking about training COCO dataset. For your case, it is different. You are training your own images but only 300 images. Although it is a bit less, you can still train with its spec and monitor the loss and AP.

Mask rcnn poor results

Accelerated Computing Intelligent Video Analytics TAO Toolkit

Morganh September 20, 2021, 1:51pm 2

For your case, the 140th iteration is the very beginning of training. Please wait for the further result during training.
More reference: Poor metric results after retraining maskrcnn using TLT notebook - #16 by ghazni

Topic		Replies	Views
Poor metric results after retraining maskrcnn using TLT notebook TAO Toolkit	23	2490	August 3, 2021
Problem in train Mask Rcnn TAO Toolkit	9	863	October 7, 2021
Maskrcnn.ipynb - followed notebook and ended up with poor (almost untrained) network from instructions TAO Toolkit	13	764	October 12, 2021
Train mask-rcnn failure TAO Toolkit tao	16	1201	November 25, 2021
Low accuracy for MS COCO dataset in tao maskrcnn model training TAO Toolkit	10	79	September 17, 2024
Nvidia TAO Maskrcnn training problem. Training is complete but Evaluation Metrics are all 0, and can not achieve right mask on inferences TAO Toolkit	6	530	May 5, 2023
Unable to visualize the mask in MaskRcnn using Nvidia tao toolkit TAO Toolkit	4	324	August 22, 2023
Training from scratch using TAO for maskrcnn TAO Toolkit	12	734	January 16, 2023
Mask_rcnn shows training logs loss 0.00000 fast_rcnn class loss: 0.00000 fast_rcnn box loss: 0.00000 TAO Toolkit	22	824	August 16, 2022
Tao mask_rcnn training exits with NaN loss TAO Toolkit	12	101	September 18, 2024

Mask rcnn poor results

Related topics