Its testcode need （torch.utils.data.DataLoader）。 you can look PSENet Project .> test_ic15.py 72 lines
I have torch==1.4.0 change PSENet.pth ==> PSENet.pt and model load in Android is OK。But，
next I don’t know what to do.
I want a little alittle translation the PSENet testcode in Android 。
Sorry，my English is very poor, if you can, give me some android advice
This Android tutorial should give you a quick overview of the basic functionality.
I’m currently unsure where you are stuck, so the tutorial(s) might be helpful.
hi, I have looked the android document when I asked。 What I want to say is ： I trained a text detection model and I wanted it to work on android. I’ve successfully loaded it on android, but the follow-up work (read pictures, output test results) I don’t have a clue. The original code used the (torch.utils.data.dataLoader) interface, so how I used it in android (torch.utils.data.data.DataLoader), I wanted to do nothing more than turn the test code step by step to android runable
Above from Bing Translation
You wouldn’t need to use a DataLoader, but could stick to the example of loading a single sample.
I’m not sure, if the DataLoader with the complete functionality is ported to the mobile API (and if not, if it’s planned to do so).
OK，thanks. I will try something