search

JaryHuang / awesome_SSD_FPN_GIoU

This repository carries out some paper recurring work
106 stars 25 forks source link

In training yourself dataest,I have two questions for you #8

Open He-XPU opened 5 years ago

He-XPU commented 5 years ago

In training yourself dataest,2.in the data/ ,you need make yourself-data.py, for example CRACK.py(it is my dataset), and change it according to your dataset. Also,in the init.py you can write something about your dataset 3.change the config/ config.py, and make your dataset config,like CRACK(dict{}) in the 2 step,how to make yourself-data.py and change it according to your dataest?and in the init.py what i write about my dataest? in the 3 step,how to change the config/config.py and how to make my dataest config?

JaryHuang commented 5 years ago

Hello,VOC is the standard Dataset, so if you want to train your-self Data,you can follow the CRACK. which is a two classes Dataset. first,you should to learn what difference between Crack.py and VOC.py, and make your A.py. second, in init.py import your A module ,A classes. Finally, in the config.py add you A classes. if you also puzzle ，you also can make you dataset like VOC, and just change the number of class and class name.

He-XPU commented 4 years ago

I have made a dataset like VOC，where am i changing the number of class and class name?Another question.Where can I find several network models for your comparison? I also want to compare them.

He-XPU commented 4 years ago

I did not find ssd+fpn and ssd+giou online. Excuse me, Can you give me these two networks?