Useless Documentation and Sample Codes ???

The shape of weights achieved by 'dumpTFWts.py is error when use python_api like 'network.add_convolutoin(...), bu I cannot find a way to reshape or transpose the weights to meet the request in tensorrt, such that the network prediction is always error, why so useless sample codes and documentation ?

By the way, the python_api is very ugly, seems written by a stupid boy …

Are you trying to work through a specific sample?

It would help if you gave a more specific description of your problem. What are you trying to do? What are you trying and what result are you getting?

Yea we know the python API isn’t perfectly Pythonic. What would you like to see done differently in the Python API?