pyprobound.table.CountTable

class CountTable(dataframe, alphabet, transliterate=None, transliterate_flanks=False, left_flank='', right_flank='', left_flank_length=0, right_flank_length=0, max_left_flank_length=None, max_right_flank_length=None, wildcard_pad=False, min_variable_length=None, max_variable_length=None)

Bases: Table[CountBatch], CountBatch

A tensor encoding of a count table with flank management.

seqs

A sequence tensor of shape \((\text{minibatch},\text{length})\) or \((\text{minibatch},\text{in_channels},\text{length})\).

Type:: Tensor

target

A count tensor of shape \((\text{minibatch},\text{rounds})\).

Type:: Tensor

left_flank

The prepended sequence.

Type:: str

right_flank

The appended sequence.

Type:: str

left_flank_length

The scored length of the left flank.

Type:: int

right_flank_length

The scored length of the right flank.

Type:: int

counts_per_round

The number of probes in each round of the count table, as a count tensor of shape \((\text{rounds})\).

Type:: Tensor

__init__(dataframe, alphabet, transliterate=None, transliterate_flanks=False, left_flank='', right_flank='', left_flank_length=0, right_flank_length=0, max_left_flank_length=None, max_right_flank_length=None, wildcard_pad=False, min_variable_length=None, max_variable_length=None)

Initializes the count table.

Parameters:

dataframe (DataFrame) – The dataframe used to initialize the count table.
alphabet (Alphabet) – The alphabet used to encode sequences into tensors.
transliterate (dict[str, str] | None) – A mapping of strings to be replaced before encoding.
transliterate_flanks (bool) – Whether to apply transliteration to flanks.
left_flank (str) – The prepended sequence.
right_flank (str) – The appended sequence.
left_flank_length (int) – The scored length of the left flank.
right_flank_length (int) – The scored length of the right flank.
max_left_flank_length (int | None) – The maximum allowed length of the prepended sequence.
max_right_flank_length (int | None) – The maximum allowed length of the appended sequence.
wildcard_pad (bool) – Whether to append a wildcard character (ex. N for DNA) to all sequences to make them the same length.
min_variable_length (int | None) – The minimum possible length of the sequences (needed if using train/test splits on variable length data).
max_variable_length (int | None) – The maximum possible length of the sequences (needed if using train/test splits on variable length data).

Methods

`batchlen`()	Batch length.
`get_setup_string`()	A description used when printing the output of an optimizer.
`set_flank_length`([left, right])	Updates the length of flanks included in sequences.

Attributes

`input_shape`	The number of elements in the length dimension.
`left_flank`	The prepended sequence.
`left_flank_length`	The scored length of the left flank.
`max_read_length`	The maximum number of finite elements in the length dimension.
`min_read_length`	The minimum number of finite elements in the length dimension.
`right_flank`	The appended sequence.
`right_flank_length`	The scored length of the right flank.
`seqs`
`target`

Non-Inherited Members

property input_shape: int: The number of elements in the length dimension.

property min_read_length: int: The minimum number of finite elements in the length dimension.

property max_read_length: int: The maximum number of finite elements in the length dimension.

get_setup_string()

A description used when printing the output of an optimizer.

Return type:: str

set_flank_length(left=0, right=0)

Updates the length of flanks included in sequences.

Parameters:

left (int) – The new length of the prepended sequence.
right (int) – The new length of the appended sequence.

Return type:

None