Tuan Truong

Hi, regarding your questions: 1. Yes. "MEAN" image vector is the mean value of all yelp image vectors. 2. "MEAN+3" is the scenario when there is a total of 4...

1. Let me correct my previous statement. The MEAN image vector is the output from the VGG16 model when we input the BGR image of values `[103.939, 116.779, 123.68]` as...

After zero-centralization, the MEAN image will become an all-zeros image with the shape of `[1, 224, 224, 3]`, which is then input to VGG16 to obtain the MEAN feature vector....

Yes. That's the process.

I would recommend using a reference of [Experiment](https://github.com/PreferredAI/cornac/blob/5caf11cffb862c304e4dcc3e0e90c8bdcdc08093/cornac/experiment/experiment.py#L26) which has access to models, metrics, and results.

You have the option to specify name of the model in its constructor.

I don’t get it. You just need a unique identifier for each model serving as the model name. Then, you just use the experiment to access to any models and...

Thanks for pointing out the interesting resources. We will try to take a look, and of course we welcome your contributions.

1. @georgeguo-cn what you did is trying to modify the BiVAE model output. As @saghiles already explained, in some scenarios, we might not want to remove recommendations that user repeatedly...

Could you give an example on why we want to build an eval method from pre-built uid/iid maps, where train/val/test datasets provided?