We want to use a custom pretrained pytorch to work with DeepStream. We do not have the training data, just the pretrained model in .pth format. How to convert it to tlt format? The pytorch model is from GitHub - NVIDIA-AI-IOT/trt_pose: Real-time pose estimation accelerated with NVIDIA TensorRT
Morganh
2
Sorry, converting pytorch custom model to tlt is not compatible.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Importing TensorRT model to TLT | 5 | 909 | May 21, 2020 | |
| Custom Models for Deepstream (TLT) | 1 | 494 | June 25, 2021 | |
| Import tlt model in python code | 3 | 1364 | September 22, 2020 | |
| TLT to TensorRT engine | 1 | 734 | October 21, 2020 | |
| transfert learning toolkit-> export model | 10 | 4016 | October 4, 2019 | |
| How to convert a pytorch model (.pt file) to a .etlt model? | 3 | 1140 | May 5, 2022 | |
| Can i use tlt model with pytorch or tensorflow? | 1 | 905 | March 26, 2021 | |
| Do I have to train my own dataset with NVIDIA TAO Toolkit to get a .tlt model? | 10 | 935 | May 16, 2022 | |
| Tensor-rt to tflite conversion | 1 | 1627 | July 13, 2021 | |
| Using TLT trained models in TF (or Keras) to run inference | 8 | 1621 | June 22, 2021 |