how to compress in my dataset

[as754770178]

When I compress model on my custom dataset, the function def setup_lrn_rate(global_step, model_name, dataset_name) raise the NotImplement error. I want to produce a function that return lr and batch numbers in normal dataset and normal module. How I set the value of nb_epoches, decay_rates ...? How to choose piecewise constant strategy or exponential decaying strategy?

[jiaxiang-wu]

You can set these hyper-parameter similar to how you train a normal/uncompressed model on your own data set, without PocketFlow.

[as754770178]

Can you give me an example? Which parameters should I set?

[jiaxiang-wu]

Before PocketFlow, have you trained some models on this data set? What is your hyper-parameter setting, or, how do you adjust the learning rate during the training process at that time?

[as754770178]

I mean that can I set some hyper-parameter so that PocketFlow does not call setup_lrn_rate function?

[jiaxiang-wu]

Almost all model compression components need this function to setup the learning rate schedule and number of training iterations. I suggest you do not skip this function. If you do want to skip this, you should modify the corresponding model compression component, and manually setup the learning rate schedule.

[as754770178]

OK，thanks