Indexing on ndarrays - use Technical Quotient

1. Remember

What dtype is returned when accessing the field 'a' in a structured array defined as np.zeros((2, 2), dtype=[('a', np.int32), ('b', np.float64, (3, 3))])?

dtype('int32')

Correct; accessing the field 'a' directly retrieves the data type specified for 'a', which is int32.

dtype('float64')

Incorrect; float64 is the data type for field 'b', not 'a'.

dtype('float32')

Incorrect; float32 is not the specified data type for any fields in the array.

dtype('int64')

Incorrect; int64 is not the specified data type for any fields in the array.

2. Understand

What happens when you access multiple fields, 'a' and 'b', using x[['a', 'b']] in a structured array x?

Returns a copy containing only those fields.
Returns a view containing only those fields.
Merges the fields into a single new field.
Raises an error due to invalid syntax.

3. Apply

Given a structured array x = np.zeros((3, 3), dtype=[('a', 'i4'), ('b', 'f8', (2, 2))]), how would you access the 'b' field and what shape would the resulting array have?

x['b'] with shape (3, 3, 2, 2)

Correct; Accessing the 'b' field of the structured array x with x['b'] returns a view of the array with the shape of the original array appended by the dimensions of the sub-array, hence (3, 3, 2, 2).

x['b'][0] with shape (2, 2)

Incorrect; While accessing the first element of 'b' with x['b'][0] would give a (2, 2) array, the question asks about accessing the 'b' field directly from x, not a single element.

x['a'] with shape (3, 3)

Incorrect; Accessing 'a' would indeed return an array with shape (3, 3), but the question asks about the 'b' field.

x[['a', 'b']] with shape (3, 3)

Incorrect; Accessing both fields [['a', 'b']] would return a combined view, but the shape would not be (3, 3); this option also does not address the specific query about the 'b' field's shape.

1. Remember

2. Understand

3. Apply

Leave a ReplyCancel Reply