chainer.functions.linear(x, W, b=None)[source]

Linear function, or affine transformation.

It accepts two or three arguments: an input minibatch x, a weight matrix W, and optionally a bias vector b. It computes

\[Y = xW^\top + b.\]
  • x (Variable or numpy.ndarray or cupy.ndarray) – Input variable, which is a \((s_B, s_1, s_2, ..., s_n)\)-shaped float array. Its first dimension \((s_B)\) is assumed to be the minibatch dimension. The other dimensions are treated as concatenated one dimension whose size must be \((s_1 * ... * s_n = N)\).
  • W (Variable or numpy.ndarray or cupy.ndarray) – Weight variable of shape \((M, N)\), where \((N = s_1 * ... * s_n)\).
  • b (Variable or numpy.ndarray or cupy.ndarray) – Bias variable (optional) of shape \((M,)\).

Output variable. A float array with shape of \((s_B, M)\).

Return type:


See also



>>> x = np.random.uniform(0, 1, (3, 4)).astype('f')
>>> W = np.random.uniform(0, 1, (5, 4)).astype('f')
>>> b = np.random.uniform(0, 1, (5,)).astype('f')
>>> y = F.linear(x, W, b)
>>> y.shape
(3, 5)