And - 1 year ago 90

Python Question

What would be the best way of converting a 2D numpy array into a list of 1D columns?

For instance, for an array:

`array([[ 0, 5, 10],`

[ 1, 6, 11],

[ 2, 7, 12],

[ 3, 8, 13],

[ 4, 9, 14]])

I would like to get:

`[array([0, 1, 2, 3, 4]), array([5, 6, 7, 8, 9]), array([10, 11, 12, 13, 14])]`

This works:

`[a[:, i] for i in range(a.shape[1])]`

but I was wondering if there is a better solution using pure Numpy functions?

Answer Source

I can't think of any reason you would need

```
[array([0, 1, 2, 3, 4]), array([5, 6, 7, 8, 9]), array([10, 11, 12, 13, 14])]
```

Instead of

```
array([[0, 1, 2, 3, 4], [5, 6, 7, 8, 9], [10, 11, 12, 13, 14]])
```

Which you can get simply with `a.T`

^{If you really need a list, then you can use list(a.T)}