Closed Shining-Tears closed 2 years ago
谢谢~ 例子中的onnx生成的engine是fp32的吧 ,在后面加上--clip-range-file <jison文件位置>能生成int8的engine吗?
然后试了一下,两个engine文件大小并没有差距很大,正常情况int8的engine不是应该比fp32的engine小很多吗?
@tuochi 麻烦了~
This issue has not received any updates in 120 days. Please reply to this issue if this still unresolved!
现在可以通过PTQ/QAT生三个文件,包括两个onnx和一个json,那么该如何变成tensorrt的文件格式呢?