> is this intended? > > np.histogramdd([[1,2],[3,4]],bins=2) > > (array([[ 1., 0.], > [ 0., 1.]]), > [array([ 1. , 1.5, 2. ]), array([ 3. , 3.5, 4. ])]) > > np.histogram2d([1,2],[3,4],bins=2) > > (array([[ 1., 0.], > [ 0., 1.]]), > array([ 1. , 1.5, 2. ]), > array([ 3. , 3.5, 4. ])) > > np.histogram([1,2],bins=2) > > (array([1, 1]), array([ 1. , 1.5, 2. ]))
maybe i should have been more explicit. what i meant to say is that 1. the counts in a histogram are integers. whenever no normalization is used i would expect that i get an integer array when i call a histogram function. 2. now it might be intended that the data type is always the same so that float has to be used to accomodate the normalized histograms. 3. in any case, the 1d histogram function handles this differently from the 2d and dd ones. this seems inconsistent and might be considered a bug. n. _______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion