[FIXED] Why doesn't scientific notation work in Numpy

Issue

Why in Ipython does:

In [1]:    1E6
Out[1]: 1000000

work but:

In[2]: import numpy as np
       rng = np.random.RandomState(42)

In[3]: rng.rand([1E6])

produces: TypeError: 'float' object cannot be interpreted as an integer

however if I type in:

In[4]: rng.rand(1000000)

It works?

Solution

Remember that scientific can also be used to represent tiny numbers:

1E-3

so using a float here makes sense. If you now check the type of 1E6 you will notice

type(1E6)  # <class 'float'>

whereas

type(1000000)  # <class 'int'>

And your code complains

TypeError: 'float' object cannot be interpreted as an integer

So if you explicitly cast 1E6 to an int it will work as expected

rng.rand(int(1E6))

Answered By - Nils Werner