lzcchl

about the Quantize layer when trans model

2

I have two model, which is mobilenetv1 for classification. the first model, it's download from google: https://storage.googleapis.com/download.tensorflow.org/models/tflite/mobilenet_v1_224_android_quant_2017_11_08.zip the second model, it's ctreat by myself, I make its layers same to...

您好，您的Pelee + SSD + driverable maps项目非常优秀，请问有没有教程来使用它呢？后期会加入车道线的检测吗？

Classification

1

I trained weights well, how to throw a picture in when evaluating, which category does it tell me to belong to?

Do you consider providing training code as it provide in https://github.com/mks0601/V2V-PoseNet_RELEASE

how can i build on windows

1

how can i build on windows

about C++ backend only gpu

3

hello, I have reviewed some materials, but there may be some omissions. Regarding a C++backend Ensemble pipeline model, how to ensure that the data flow always stays on the GPU...

in models/random-tree-walks/rtw.py, after train model and testing, you write "y_pred[:, :, 2] = y_test[:, :, 2]", you make truth Z copyTo pred, so can I understand that this model can't...

why did you get 3d groundtruth by using opencv sfm instead of depth camera itself?

1

at first, It is very simple for everyone to obtain the 3d coordinates of the human or other things surface points; secondly, as I understand, opencv sfm is a method...

model analyzer: profiling of multiple models is not supported for ensemble models

2

version: tritonserver:23.12-py3-sdk In my case, I have some ensemble models, I want to know how to set the parameters so that my device can achieve optimal performance (maximum throughput). but...

How to convert functional.Tensor to numpy or a list of Tensor ?

1

Due to upgrading the version of trtllm from 0.8.0 to 0.9.0, I originally obtained the "host_max_attention_window_sizes" as a List in models/generation_mixin.py, but now it is a Tensor, so in order...