It is very important for training to record the parameters or data!

Code

	import numpy as np
	# define
	dict = {'a' : {1,2,3}, 'b': {4,5,6}}
	# save
	np.save('dict.npy', dict)
	# load
	dict_load=np.load('dict.npy', allow_pickle=True)
	
	print("dict =", dict_load.item())
	print("dict['a'] =", dict_load.item()['a'])

output:

dict = {'a': {1, 2, 3}, 'b': {4, 5, 6}}
dict['a'] = {1, 2, 3}

References

• numpy.load
• numpy.save

Note

numpy.savez is also a interesting method, you could use it to record the specific array with keywords like dict. The type of file is .npz

• Example

>>> from tempfile import TemporaryFile
>>> outfile = TemporaryFile()
>>> x = np.arange(10)
>>> y = np.sin(x)

>>> np.savez(outfile, x=x, y=y)
>>> npzfile = np.load(outfile)
>>> npzfile.files
['x', 'y']
>>> npzfile['x']
array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])

from numpy.savez