I'm thinking that any one of several light US aircraft might fit this. The Ryan PT-22 with the Kinner is one, but there is also the Aeronca LC and the Culver Dart, powered by Warner, LeBlond, or Ken Royce 5 cylinder radials. The reason for the odd number of cylinders in a single row radial is that there is a single throw on the crankshaft, and a four stroke radial will make two revolutions to fire all of its cylinders. It's easier to get the valve timing and ignition timing done mechanically by firing alternate cylinders. 1,3,5,2, 4 is the firing order. The 3 cylinder radial that was most common, even though it was pretty much a disaster, was the Szekely. One of my uncles learned to fly in a Curtiss Junior powered by the Szekely, and he had a cable attached to all of the cylinder heads to keep them from going through the pusher prop when they broke. I kid you not.
There was a six cylinder 4 stroke radial, the Curtiss Challenger, and it has fooled some people because there was no main bearing between the two throws of the crankshaft and at first glance it can appear that all six of the cylinders are in line, but they are not. It is a twin row 3 in front 3 in back radial. It wasn't a good engine because of the lack of the middle main, but it did power some Curtiss Robins.
It is possible to have 5 cylinders in an inline 4 stroke engine, Mercedes had a diesel with 5 cylinders. I suspect it had irregular timing, like the Wisconsin V-4 industrial engine that used the same crank as an inline engine. Two cylinders fired 180 degrees apart as is normal, but the other two were 90 and 270 degrees in interval. It was not a smooth running engine.
With two stroke engines, any configuration is possible. Detroit diesel sold I3, I4, I6, V8 and V12 cylinder versions of the V-71 series diesels.