周周周

[Paddle-TRT] fix conv2d

1

### PR types Others ### PR changes Others ### Describe When filter is not a persistable tensor, forbid it entering into trt

[Paddle-TRT] constnt-folding

1

### PR types Others ### PR changes Others ### Describe Others

### PR types Others ### PR changes Others ### Describe - conv2d_transpose至少从7134开始就支持不对称的padding，可认为其一直支持 - conv3d_transpose自8406开始支持不对称的padding了

Inferred shape and existing shape differ in dimension 0: (1) vs (-1)

1

# Bug Report ### Is the issue related to model conversion? no ### Describe the bug when I use `shape_inference.infer_shapes` to infer my onnx model, it reportes Inferred shape and...

Scale: scale (Output: tmp_131142): only activation types allowed as input to this layer

2

## Description ## Environment **TensorRT Version**: 8405 **NVIDIA GPU**: **NVIDIA Driver Version**: **CUDA Version**: **CUDNN Version**: **Operating System**: **Python Version (if applicable)**: **Tensorflow Version (if applicable)**: **PyTorch Version (if applicable)**:...

addFullyConnected will be replaced by addMatrixMultiply + addElementWise?

1

## Description ## Environment **TensorRT Version**: **NVIDIA GPU**: **NVIDIA Driver Version**: **CUDA Version**: **CUDNN Version**: **Operating System**: **Python Version (if applicable)**: **Tensorflow Version (if applicable)**: **PyTorch Version (if applicable)**: **Baremetal...

why 7234 和 8406 diffs in layer's output dimensions when constructing network

3

## Description ## Environment **TensorRT Version**: **NVIDIA GPU**: **NVIDIA Driver Version**: **CUDA Version**: **CUDNN Version**: **Operating System**: **Python Version (if applicable)**: **Tensorflow Version (if applicable)**: **PyTorch Version (if applicable)**: **Baremetal...

[Paddle-TRT][Cherry-Pick]Trt reshape2 matmul fuse pass 2 4

1

### PR types Others ### PR changes Others ### Describe [Cherry-Pick](https://github.com/PaddlePaddle/Paddle/pull/46090)

[Paddle-TRT]Fix cast bug

1

### PR types Others ### PR changes Others ### Describe layer->getOutput()->setType() may fail, use layer->setOutputType() instead. in trt 8.4 setType fails， but ok in trt 8.2, so strange，so I use...