Functions — github.com/Blealtan/efficient-kan

CodeHub MCPcopy Create free account

hub / github.com/Blealtan/efficient-kan / functions

Functions13 in github.com/Blealtan/efficient-kan

⨍ Functions 13 ◇ Types & classes 2

↓ 3 callersMethodb_splines

Compute the B-spline bases for the given input tensor. Args: x (torch.Tensor): Input tensor of shape (batch_size, in_fea

src/efficient_kan/kan.py:78

↓ 2 callersMethodcurve2coeff

Compute the coefficients of the curve that interpolates the given points. Args: x (torch.Tensor): Input tensor of shape

src/efficient_kan/kan.py:113

↓ 1 callersMethod__init__

( self, layers_hidden, grid_size=5, spline_order=3, scale_noise=0.1,

src/efficient_kan/kan.py:241

↓ 1 callersMethodregularization_loss

(self, regularize_activation=1.0, regularize_entropy=1.0)

src/efficient_kan/kan.py:281

↓ 1 callersMethodreset_parameters

src/efficient_kan/kan.py:56

↓ 1 callersMethodupdate_grid

(self, x: torch.Tensor, margin=0.01)

src/efficient_kan/kan.py:169

( self, in_features, out_features, grid_size=5, spline_order=3,

src/efficient_kan/kan.py:7

Functionclosure

tests/test_simple_math.py:15

(self, x: torch.Tensor)

src/efficient_kan/kan.py:153

(self, x: torch.Tensor, update_grid=False)

src/efficient_kan/kan.py:274

Methodregularization_loss

Compute the regularization loss. This is a dumb simulation of the original L1 regularization as stated in the paper, since t

src/efficient_kan/kan.py:217

Methodscaled_spline_weight

src/efficient_kan/kan.py:146

Functiontest_mul

tests/test_simple_math.py:8