Open Four1996 opened 2 years ago
now i solved it with smaller input size
Hello @Four1996
I'm training on a custom dataset and it seems that it costs nearly a week to complete 80 epochs. I just modified the annotation file path in yolox_x_mix_det.py
for my exp, and the training data consists of 42750 imgs with 1920x1080.
How did you set the input size? The time is just too long for me.
My device: Quadro RTX 8000 x8
In my log the data_time
can be up to 30s, I guess that may be the reason. I'm trying to solve it.
Having moved the data to SSD, the problem has been solved.
I use
python3 tools/train.py -f exps/example/mot/yolox_x_ablation.py -d 3 -b 8 --fp16 -o -c pretrained/yolox_x.pth
inTrain ablation model (MOT17 half train and CrowdHuman)
,got too long training time. My device: RTX 2080 ti x3, Intel(R) Xeon(R) CPU E5-2650 v4 @ 2.20GHz x4。 Here is my log file: