ipython3 numpy数组未保存

Question

I have a Filter object containing an array with data, and when iterrating over the object and trying to save the data.. it zeros out and nothing is getting saved.

Notebook Code

valid = filter(None, results)
print(results)
filenames = [x[0] for x in valid]
samples = [x[1] for x in valid]
durations = [x[2] for x in valid]
print(samples)
samples = np.asarray(samples)
np.savetxt(join(data_root, 'filenames.txt'), filenames, fmt='%s')
np.savetxt(join(data_root, 'durations.txt'), durations, fmt='%i')
%time np.save(join(data_root, 'samples.npy'), samples)
print('Saved', len(list(valid)), 'samples')

Output on running in Jupyter Notebook

[('data/drums/samples/anywhereyougo_1.2sec.wav', array([-0.00331263, -0.00248447,  0.00124224, ...,  0.        ,
   -0.00207039, -0.00455487], dtype=float32), 58576), ('data/drums/samples/anywhereyougo_2.8sec.wav', array([ 0.        ,  0.00041356,  0.        , ...,  0.        ,
    0.00041356, -0.00041356], dtype=float32), 136053)]

[]
CPU times: user 362 µs, sys: 205 µs, total: 567 µs
Wall time: 612 µs
Saved 0 samples

And all I get in the samples.npy is the following :

**�NUMPYF{'descr': '<f8', 'fortran_order': False, 'shape': (0,), }            **

I'm assuming that's an empty num py array.. just started with numpy

Answer 1

So after looking at the original code in a non-iPython interpreter

rearranged things :

samples = []
filenames = []
durations = []
sample_total = 0

valid = filter(None, results)
for x in enumerate(valid):
    filenames.append(x[1][0])
    samples.append(x[1][1])
    durations.append(x[1][2])
    sample_total = x[0] + 1

samples = np.asarray(samples)

np.savetxt(join(data_root, 'filenames.txt'), filenames, fmt='%s')
np.savetxt(join(data_root, 'durations.txt'), durations, fmt='%i')
%time np.save(join(data_root, 'samples.npy'), samples)

print("Saved " + str(sample_total) + " samples")