1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
|
"""
# -*- coding: utf-8 -*-
#
# Copyright 2021 Michael Büsch <m@bues.ch>
#
# Licensed under the Apache License version 2.0
# or the MIT license, at your option.
# SPDX-License-Identifier: Apache-2.0 OR MIT
#
"""
from mlplib.activation import *
from mlplib.backward import *
from mlplib.forward import *
from mlplib.gradient_check import *
from mlplib.init import *
from mlplib.loss import *
from mlplib.parameters import *
import numpy as np
def test_backward_prop():
seed(42)
inputs = 4
layout = (6, 6, 9, 2)
params = Parameters(
weights=init_layers_weights(inputs, layout),
biases=init_layers_biases(layout),
actvns=[
ReLU(),
LReLU(0.1),
Tanh(),
Sigmoid(),
],
)
x = standard_normal((20, inputs))
y = standard_normal((20, layout[-1]))
gradients, yh = backward_prop(x, y, params, MSE())
yh2 = forward_prop(x, params)
assert np.all(yh == yh2)
ok = gradient_check(x, y, params, MSE(), gradients)
assert ok
# vim: ts=4 sw=4 expandtab
|
