lantudou

Hi, I am a freshman of this library, I want to use some transformation which could lead to the change of bbox location and range to argument my dataset of...

Got very high recall but low precision in my own two-classes dataset

1

Hi, I tried to train the original yolov4 model on my own datasets. Why I will got a model with very high recall but very low precision? , It seems...

A bug for plotting my own data

1

I used your code to plot my own data and find out an abnormal grey rectangular in the plot.  Using the following command: `vs = violinplot(appliance_data);` for this data:...

About high cuda memory allocation with too long context length

1

In training process, I have noticed that the cuda code finished all calculation within the ctx-len, the speed is fast but seems memory-unfriendly for some application with long context length....

The validation result is lower than the training accuracy using selsa model for own dataset

2

Hi, I want to train the selsa model for my own dataset. A sample of my customized dataset json is given below: `{"categories": [{"id": 1, "name": "car"}], "images": [{"id": 1,...

Does anyone have a good reproduction result？

2

Tried to train a model using sequence dataset with provided virtual data, and I found very poor performance and different loss recording in log. Could anyone give me more detail...

when use own defined convertor, why int tensor input has no _trt attribute?

2

here is a simple reproduce code ``` import os import time import torch import torchvision from torch2trt.torch2trt import * import numpy as np class Sparse(torch.nn.Module): def __init__(self, embedding_size): super().__init__() self._embedding_size...