How should I varify the trained model by run_val.sh?

[yushanshan05]

I have seen the code that the validated datavaridat is set by the parameter of ‘--data-val'. And the default value is '/tmp/val.rec‘. But I don't know what and where this file is. And if I want to test a image, how should I do? Thanks very much!

[jiangqr]

use im2rec can generate rec file

[cypw]

@yushanshan05 @whcacademy

You should be able to fully reproduce the result by the following steps: ( I've double checked by myself )

1. Download ImageNet-1k validation set
2. git clone --recursive https://github.com/cypw/DPNs.git
3. Setup DPNs/mxnet && make
4. Generate the val.rec file by $im2rec_path $val_lst_path $val_lst_pref ./val.rec unchanged=1
5. Set the path for val.rec in scores.py
6. run ./run_val.sh

[hcwang95]

That's cool! Thanks for your help!

[gjd2017]

Hi, I have trained the model(Dpn-92) based on my own dataset, the epoch is 20, however, when I use scores.py and run_val.sh test my dataset, there is a mistake, the val.log is empty. How should I do? Thanks very much!

[gjd2017]

Solved