Allow expanded physical signal in calc_adc_params

[briangow]

This PR updates calc_adc_params to allow an expanded physical signal to be passed. Previously only a non-expanded signal was allowed. The function will now calculate the adc_gain and baseline for p_signal or e_p_signal.

I check for the signal that is not None and then compute the gain and baseline. I moved the calculation itself to a new function calc_adc_gain_baseline so we could call it for either signal type.

[bemoody]

Thanks!

It could be optimized somewhat: calculating min/max using a generator expression

minvals.append(min(x for x in ch if not math.isnan(x)))

will be slower than doing it with numpy. But the important thing is that it works.

It would be good to add a test case; currently it looks like there is no test coverage for calc_adc_params. The easiest way to test it is using wfdb.wrsamp. So we should do that as a follow-up after merging this and #510.

[briangow]

Thanks!

It could be optimized somewhat: calculating min/max using a generator expression

minvals.append(min(x for x in ch if not math.isnan(x)))
will be slower than doing it with numpy. But the important thing is that it works.

It would be good to add a test case; currently it looks like there is no test coverage for calc_adc_params. The easiest way to test it is using wfdb.wrsamp. So we should do that as a follow-up after merging this and #510.

@bemoody , I have updated the code to use np.nanmin and np.nanmax instead of the generator function, as suggested.

I can look into adding a test once these PRs are merged.

[bemoody]

This looks good!