API Reference

`unpack_to_array(qa, product, flags='all')`

Make a mask array with the same shape as qa with an additional axis

for all flag masks.

Parameters:

Name Type Description Default

qa

np.array or int

An array of any shape with an int-like dtype. If a single integer it will be coverted to a numpy array with length 1.

required

product

str or dict

A unique product identifer string. See list_products() for availability.

A custom bit flag specification may also be used via a dictionary. For example, below is an 8 bit qa flag where the 1st and 2nd flags are bits 0 and 1, and the 3rd flag is spread across bits 4 and 5. Note bit 3 is not specified so is ignored. max_value is generally set to the maximum possible value given the bit size. If this option is used then the flags argument can be left as the default all.
Flags should be ordered from lowest to highest bits. The order of the mask axis in the returned array will be the same as the flag order specifed here.

product_info = {'flag_info':{'flag1_description':[0],
                             'flag2_description':[1],
                             'flag3_description':[4,5]},
                'max_value' : 255,
                'num_bits'  : 8}

required

flags

list of strings or 'all'

List of flags to return. If 'all', the default, then all available flags are returned in the array. See available flags for each product with list_qa_flags().

'all'

Returns:

Type	Description
`np.array`	If only a single flag is set then the array shape will be the same as qa.shape. If > 1 flag is set then an axis will be added in position -1, with the shape as qa.shape + (n_flags,) Ordering of the flag axis will be the same order of the `flags` list argument. If `flags` is 'all' then the order is the same as that in `available_qa_flags`, which are be aligned with the flag bit order.

`unpack_to_dict(qa, product, flags='all')`

Get mask arrays for the specified flags in a dictionary format.

Parameters:

Name Type Description Default

qa

np.array or int

An array of any shape with an int-like dtype. If a single integer it will be coverted to a numpy array with length 1.

required

product

str or dict

A unique product identifer string. See list_products() for availability.

A custom bit flag specification may also be used via a dictionary. For example, below is an 8 bit qa flag where the 1st and 2nd flags are bits 0 and 1, and the 3rd flag is spread across bits 4 and 5. Note bit 3 is not specified so is assumed unused. max_value is generally set to the maximum possible value given the bit size. If this option is used then the flags argument can be left as the default all.
Flags should be ordered from lowest to highest bits. The resulting dictionary key names will be the same as the flag descriptions specified here.

product_info = {'flag_info':{'flag1_description':[0],
                             'flag2_description':[1],
                             'flag3_description':[4,5]},
                'max_value' : 255,
                'num_bits'  : 8}

required

flags

list of strings or 'all'

List of flags to return. If 'all', the default, then all available flags are returned in the array. See available flags for each product with list_qa_flags().

'all'

Returns:

Type	Description
`dict`	A dictionary where the flag names are keys and values are np.arrays of the flag mask with shape qa.shape. All entries will have the same shape.

`list_products(sensor='all')`

Return a list of available products for QA unpacking.

'all', the default, returns the identifers for all products for all sensors

Parameters:

Name	Type	Description	Default
`sensor`	`int`	Sensor ID. The default is 'all'.	`'all'`

Returns:

Type	Description
`list`	List of product identifers

`list_qa_flags(product)`

Return a list of QA flags configured for a specific product.

Parameters:

Name	Type	Description	Default
`product`	`str`	A unique product identifer. See `list_products()` for availability.	required

Returns:

Type	Description
`list`	List of QA flags. The descripters are used in the flag argument of `unpack_to_array()` and `unpack_to_dict()`.

`list_sensors()`

Return a list of available sensors configured in the package.

Returns:

Type	Description
`list`	List of sensor names

`InvalidProductSpec`

Error class for when an invalid custom specification is passed to the product argument.