fatih akyon

@lvwerra do you have any opinion on that?

@mristin @junjuew you can also utilize this [easy-to-use & packaged version of the latest yolov5 library](https://github.com/fcakyon/yolov5-python) for this project.

@muellerzr thanks for the explanation. I have one question: Why can't Neptune log function can be implemented as: ```python def log(self, values: dict, step: Optional[int] = None, **kwargs): """ Logs...

Then maybe I can test it locally and once I am sure I can work on a PR if you are OK with it :)

@kadirnar soyle bir hata aliyorum video_predict yampaya calisirken: ``` Sizes of tensors must match except in dimension 1. Expected size 46 but got size 45 for tensor number 1 in...

> > @kadirnar soyle bir hata aliyorum video_predict yampaya calisirken: > > ``` > > Sizes of tensors must match except in dimension 1. Expected size 46 but got size...

@multimeric here is the pretrained weights: https://drive.google.com/drive/folders/0BwbIUfIM1M8sc0tEMGk1NGlKZTA?resourcekey=0-oRKU3d5c8GsGm0Er0eLIjg

@yjdeveloper @snphnolt use this version with -map 0.02 for map calculation at every 0.02 epoch (starts after warmup iterations) https://github.com/fcakyon/darknet

@chekoduadarsh @hansheng0512 try the steps in this file: https://github.com/artest08/LateTemporalModeling3DCNN/blob/6f86cccedee0851a2c54f9a0bca096626095ae88/requirements.txt

@annalename thanks for contributing to this repo! Would you provide some more detail? Your Python version, torch/tensorflow version, CUDA version and the way of your CUDA installation (from cuda or...