Calling astype on a Data object with singleton entries yields an error. To reproduce, use:

data = tsc.loadExample('fish-series')
data.seriesMean().astype('int').first()

The error is TypeError: astype() takes no keyword arguments. It appears that when calling astype on a numpy.float64 (or other scalar), it doesn't take the same any additional arguments that work when calling it on an array. For example,

x = np.array([1.0,2.0,3.0])
x.astype('int', casting='unsafe')

works fine but this doesn't

x = np.float64(1.0)
x.astype('int', casting='unsafe')

One potential solution would be to check the length of the index before calling astype, and to only include the additional arguments if it's greater than 1.