Better explanation on testing

[kukuruza]

Hi, first of all, thanks for the great work! It was really easy to use. I had a problem with refining my images with a pretrained model though.

In readme you say: "To refine all synthetic images with a pretrained model":

$ python main.py --is_train=False --synthetic_image_dir="./data/gaze/UnityEyes/"

You are missing the load_path argument. Apparently, it is path to the directory with models, and it is relative to logs dir. For example:

$ python main.py  --is_train=False --synthetic_image_dir="./data/gaze/UnityEyes/" --load_path generative_2017-03-07_01-40-07

There is no indication whatsoever, that the model is loaded rather than initialized during testing. And of course if it is initialized then it will simply write garbage to refined images, leaving you wondering.

I don't know though how to make it clear whether the model was loaded or initialized if you use tf.train.Supervisor, maybe specify wait_for_checkpoint=True when calling prepare_or_wait_for_session.

[carpedm20]

I see. Yes, I forgot to add load_path. In terms of initialization or loading, we can try something like https://github.com/openai/universe-starter-agent/blob/master/worker.py#L45. If you want you can give me an PR.

[luochuwei]

@kukuruza Have you solved the problem? Is the model loaded during testing ?

It seems that when we test our own images, the model wasn't loaded. And I could not get well refined images.

[kukuruza]

@luochuwei, nothing beyond what I wrote in the first message. You need to add --load_path argument, where you specify the path to directory with your models relative to logs directory. E.g. --load_path generative_2017-03-07_01-40-07.