How to retrain a pruned model in classification?
i.e Which spec configuration need to change?
Please refer to the notebook inside the docker. The re-trained spec is also available at each notebook.
Where is the Jupyter notebook file you’re referring to? Can you refer me to any documentation about this and/or other notebooks in the Docker container? Thanks in advance…
See details in https://docs.nvidia.com/metropolis/TLT/tlt-getting-started-guide/index.html#building_docker for triggering docker and Jupyter notebooks.
There are four notebooks: classification, detectnet_v2, faster_rcnn and SSD.
The clssification network’s spec locates at /workspace/examples/classification/specs
Thanks, Morgan, very helpful!