beatmup.nnets.Dense

class beatmup.nnets.Dense

Dense (linear) layer. Computes A*x + b for input feature vector x, a matrix A and an optional bias vector b. Accepts a GL::Vector or a flat Storage view on input, amd only a GL::Vector on output.

__init__(self: beatmup.nnets.Dense, context: beatmup.Context, name: str, num_output_dims: int, use_bias: bool) → None

Creates a Dense operation.

Parameters:

  • context – a context instance

  • name – operation name

  • num_output_dims – number of output dimensions

  • use_bias – if True, the bias vector addition is enabled

Methods

__init__(self, context, name, ...)

Creates a Dense operation.

Attributes

bias_chunk_suffix

input_count

Number of operation inputs

matrix_chunk_suffix

name

Operation name

output_count

Number of operation outputs