There is something curious here. The second flush() fails. Can anyone explain this?
Tom ------------------- code snippet ------------------------ ... # create a memmap with dtype and shape that matches the data fp = np.memmap(filename, dtype='float32', mode='w+', shape=(3,4)) print "Initial memory mapped array (mode 'w+'):\n",fp # write data to memmap array fp[:] = data[:] fp.flush() # append a row to the array fp = np.append(fp, [[12,13,14,15]], 0) print "Filled memory mapped array:\n",fp fp.flush() ... ----------------- output ----------------------- Filled memory mapped array: [[ 0. 1. 2. 3.] [ 4. 5. 6. 7.] [ 8. 9. 10. 11.] [ 12. 13. 14. 15.]] --------------------------------------------------------------------------- AttributeError Traceback (most recent call last) ..../memmap_test.py in <module>() 21 fp = np.append(fp, [[12,13,14,15]], 0) 22 print "Filled memory mapped array:\n",fp ---> 23 fp.flush() 24 25 # deletion flushes memory changes to disk before removing the object AttributeError: 'numpy.ndarray' object has no attribute 'flush' WARNING: Failure executing file: <memmap_test.py> _______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion