Thiago Matheus

I mean, there isn't such a way of adding the .py files or .sh files before writing the image? Like when creating the block, you upload the file? For what...

Problems on converging ResUneta-d6 multitasking

28

Hello again @feevos, I am using Tanimoto dual loss with complement and it's working with simple ResUnet-a. However, when training on multitasking using Tanimoto on all tasks, only the first...

How do I use 2 gpus to train the same yolov3 model?

8

I'm having trouble training my yolov3 in my PC, which has 2 gpus of 8 GB. The problem is I can only train it setting Batch = 64 and Subdivisions=64....

How do you use TLT toolkit to train MaskRCNN since TLT toolkit does not support MaskRCNN model???

1

TLTtoolkit just supports ssd, fasterRCNN, and DetectNet_v2

the SSD sample depends on the batchTilePlugin. To use these plugins for the samples here, complile a new libnvinfer_plugin.so* and replace your system libnvinfer_plugin.so*

Nothing being segmented in task coronary_arteries

2

Does anyone have this problem? The model is outputting a model of zeros when using the task coronary_arteries. I tried using both CCAT and CT thorax images. I don´t know...