-
Hi!
I am compressing a pytorch DB model (like [this](https://arxiv.org/abs/1911.08947)) with quantization algorithm. I do it strictly as it is specified in the documentation and examples. After the …
-
## Description
First, I use the training data set to train a model, and the mAP of the verification set is 0.65. Then, refer to the tutorial in the Tensorrt document, and replace the conv in the mo…
-
## Description
How to solve `resize(upsample)` op in int8 by QAT(tools/pytorch-quantization) ? Except use `ConvTranspose `
## Environment
**TensorRT Version**:8.4
-
### Describe the bug
Have tried out all the models and can't get over 20 fps with the predict_webcam function on a RTX4090 GPU. For comparison the yolov8 models run at 100-120 fps.
### Video
ht…
-
你好,我想请教下,lsq的scale和zeropoint的反向传播的梯度更新在哪实现的,
我看代码中只写了forward和symbolic,但是没有实现backward
-
I believe EqualizeGraph should be considered as a separate, optional step when doing PTQ.
For some models, it just fails because of mismatched dimensions.
In other cases, it does not fail even tho…
-
## Bug Description
When using the PyT-QAT toolkit, QAT perf is slower than PTQ, for TRT this is not the case.
Torch-TRT:
Model | Accuracy | Performance
-- | -- | --
Baseline MobileNetv2 | …
-
Does tensorrt have any optimization for vision transformer deployment on AGX Orin? Int8 doesn't seem to work if I don't put a lot of parts into .cu?
-
你好。
我运行了application/imagenet_example/PTQ下的demo,尝试对resnet18进行后量化,backend为SNPE,量化方法为Adaround。最后生成的res18_clip_ranges.json中为什么权重量化参数param_encodings是空的?
-
多输入为有不同的shape,2/3/4维都有,请问这个QAT范式支持吗?如果支持的话,在保存模型前的一次前向推理应该怎么写呢?如官方给出的example:
predict = net.forward(F.placeholder([1, 3, 224, 224], F.NC4HW4))
那么请问我这个多输入的应该怎么写吗?我尝试按照net的输入顺序一次填入,执行失败了(没有报错),结果出现很…