Warren Weckesser wrote:
Benjamin Root wrote:
Note that this isn't restricted to structured arrays. For regular ndarrays, loadtxt() appears to mimic the behavior of np.squeeze():
Exactly. The last four lines of the function are:
X = np.squeeze(X) if unpack: return X.T else: return X
It would be useful to be able to tell loadtxt to not call squeeze, so a program that reads column-formatted data doesn't have to treat the case of a single line specially.
I agree -- it seem to me that every time I load data, I know what shape I expect the result to be -- I'd never want it to squeeze. It might be nice if you could specify the dimensionality of the array you want. But for now: can you just do a reshape? In [42]: strData = StringIO("53.2 49.2") In[43]:a=p.loadtxt(strData,dtype=[('x',float),('y',float)]).reshape((-1,)) In [45]: a.shape Out[45]: (1,) -Chris -- Christopher Barker, Ph.D. Oceanographer Emergency Response Division NOAA/NOS/OR&R (206) 526-6959 voice 7600 Sand Point Way NE (206) 526-6329 fax Seattle, WA 98115 (206) 526-6317 main reception Chris.Barker@noaa.gov