Xin Lin
Xin Lin
![image](https://user-images.githubusercontent.com/41569547/179931957-ccbe12bd-f591-43c1-9560-dd6a4d68753f.png) Does anyone come across the same problem? For the batchsize of 1, the coversion is ok. Does that mean nanodet does not support dynamic inference by using tensorrt?
I trained my model using based on the instructions and got really great results, but I found that the model still used GPU taking up about 1.8G to make predictions....
I am bit confused about the final result by running the main.py when inferencing. Are they just a couple of feature embeddings for the input images?
I am a bit confused with the word 'offline'. Does that mean I do not have to log into the ngc registry by using this replicator to run the pre-trained...