Cliff Hansen

I edited the checklist to defer PRs that aren't quite ready (either need reviews or author responses). The only one to decide is #1478

> I can see errors in the next file: "test.forecast.py". What can we do in this case? You can ignore that. Please assign me to review when this PR is...

Keep in mind, the units on the Boltzmann constant are different : `sapm` uses J/K, `calcparams_desoto` uses eV/K.

`EgRef` and `dEgdT` are model parameters specific to De Soto and CEC. I wouldn't put them in the constants module, although `EgRef` has a physical basis, its meaning in these...

Agree, CONSTANTS for v0.9

I see no reason not to use `del`. Perhaps your aversion is a sense that needing to use `del` is a clue that the algorithm has room for improvement. What...

The only legacy interface for loss models is in `ModelChain.losses_model`, which sits between `ModelChain.dc_model` and `ModelChain.ac_model` in the various `run_model` methods. The current options for `losses_model` are `PVSystem.pvwatts_losses` and `ModelChain.no_extra_losses`....

> Or perhaps instead of renaming `losses_model`, it could be kept as a `ModelChain` attribute where a user could specify it to be either 'pvwatts' or 'pvsyst'? Then instead of...

@ColmBhandal thanks for this report. As author of that function, I am 99.99% certain that I was not mindful of the distinction between `array_like` and `array-like`; most likely, it was...

I'm coming around to the view that it may be better to just list the specific types in the docstring, e.g., ``` x : scalar, np.array, or pd.Series ``` rather...