An easy to use PyTorch to TensorRT converter
torch.nn.functional.group_norm
using native TensorRT layerstorch.nn.ReflectionPad2d
using plugin layerThis version includes the introduction of the Quantization Aware Training workflow in torch2trt.contrib (thanks to @SrivastavaKshitij). It also contains various converters added since the previous release. Please see the notes below.
torch.nn.functional.adaptive_avg_pool3d
torch.nn.functional.adaptive_max_pool3d
torch.maxpool3d
and torch.nn.functional.max_pool3d
torch.roll
torch.nn.functional.layer_norm
torch.nn.functional.gelu
torch.nn.functional.linear
torch.nn.functional.silu
torch.Tensor.expand
torch
modulefloordiv
, mod
, ne
, and torch.tensor
operationsrelu
converter to support Tensor.relu
operationsigmoid
converter to support Tensor.sigmoid
operationTRTModule
classtorchvision
image classification models