Lim Zhen Xian

Results 17 comments of Lim Zhen Xian

> Do you recommend any other yolov4 repo compatible with pytorch? u may try wongkiuyiu version, I tested its working.

> too slow for searching the image labels in JSON file, around 10 minutes for each image. > > by the way, there are 2 bugs in `evaluate_on_coco.py` (one is...

> I changed the `conf_thresh` to `0.5` in the `do_detect()` call, so this time it won't generate several thousands of bounding boxes from `do_detect()` function. But the evaluation only does...

> > remove that 0.4 ( confidence ) and it will work but you will get another error regarding unknown image type > > I don't think you should remove...

> Do you provide train.txt which contains image paths in your data.yaml? ![image](https://user-images.githubusercontent.com/24875124/169975223-f51653f8-a66d-4cb0-9943-efc4486421ab.png) but still cannot detect my label, why is this happened? @WongKinYiu

我知道你也很久没关注这个repo了 纯粹试试运气看楼主大神会回复没有罢了,如果您有看到,请回复不管多久了 我都会等的 哈哈哈哈哈哈哈哈哈哈哈 (不是在告白)

哇哇哇 没想到你会回复而且还那么快!然后,erm, 我不是在train.py出问题勒, 我是在test.py出现这个问题。 然后刚刚我试完了,是8个epoch或以下都行但是到9个以上时候就会出现这个 问题了。

然后请问下面那个link是来做什么的? 对不起我是个objectdetection的 小白。 然后可能你也需要知道的是,我是用你的repo来训练我在的训练集的,有关系到吗?

也就是说可能我下载你的这个file,然后放进coco里面可能就行了? 我试试

你好,我是来自马来西亚。刚刚我google了下 发现你来自台湾。我想请问您这个是怎么用的?我下载了但是没有照片?