lightgbm.Sequence

class lightgbm.Sequence[source]

Bases: abc.ABC

Generic data access interface.

Object should support the following operations:

# Get total row number.
>>> len(seq)
# Random access by row index. Used for data sampling.
>>> seq[10]
# Range data access. Used to read data in batch when constructing Dataset.
>>> seq[0:100]
# Optionally specify batch_size to control range data read size.
>>> seq.batch_size
  • With random access, data sampling does not need to go through all data.

  • With range data access, there’s no need to read all data into memory thus reduce memory usage.

New in version 3.3.0.

batch_size

Default size of a batch.

Type

int

__init__()

Methods

__init__()

Attributes

batch_size