Nikolay Nez

> To make it easier to run it on our accelerator, I'm planning following experiments > > - replace stride=2 conv with max pooling or space_to_depth > - remove 5x5...

> -Timing is now only done when specified, not always by default. Debug options should not be the default... Actually, the only uses of `lm_fpga` are - testing - time...

If "extenal" time measurements imply measurement of the execution time of whole binary itself, they do not make much sense to me, as there are lots of "preparation steps" like...

As for #319 documentation should say how to 1. obtain `npy`s from training checkpoint 2. convert `pb` file 3. do `make lm_fpga` 4. run `lm_fpga.elf` with the `npy`s and read...

> It does make sense to measure all loadings and inits for a program that is meant to be used by customers Ok, this is a **test** program, for developers,...

> On a side note, do we measure the time of things like `Network::init()` separately? Yes we do. You will see both init time and run time in the report....