Open sparsh-b opened 4 years ago
you need to use TLT to re-tranin the model, please refer to tlt doc https://developer.nvidia.com/transfer-learning-toolkit or ask questions in tlt forum - https://forums.developer.nvidia.com/c/accelerated-computing/intelligent-video-analytics/transfer-learning-toolkit/17
The model file provided for SSD uses resnet-18 as backbone. I want to use Mobilenet_v1 as backbone. Please let me know how to do the same.