93 lines
3.4 KiB
Python
93 lines
3.4 KiB
Python
import sys
|
|
|
|
import os
|
|
|
|
# Concat top Level dir to system environmental variables
|
|
sys.path += os.path.join('..', '.')
|
|
|
|
from util import *
|
|
from experiment import *
|
|
from network import *
|
|
from soup import prng
|
|
|
|
import keras.backend
|
|
|
|
|
|
from statistics import mean
|
|
avg = mean
|
|
|
|
|
|
def generate_fixpoint_weights():
|
|
return [
|
|
np.array([[1.0, 0.0], [0.0, 0.0], [0.0, 0.0], [0.0, 0.0]], dtype=np.float32),
|
|
np.array([[1.0, 0.0], [0.0, 0.0]], dtype=np.float32),
|
|
np.array([[1.0], [0.0]], dtype=np.float32)
|
|
]
|
|
|
|
|
|
def generate_fixpoint_net():
|
|
#NOTE: Weightwise only is all we can do right now IMO
|
|
net = WeightwiseNeuralNetwork(width=2, depth=2).with_keras_params(activation='sigmoid')
|
|
# net = AggregatingNeuralNetwork(aggregates=4, width=2, depth=2).with_keras_params(activation='sigmoid') # I don't know if this work for aggregaeting. We don't actually need it, though.
|
|
net.set_weights(generate_fixpoint_weights())
|
|
return net
|
|
|
|
|
|
def vary(old_weights, e=1.0):
|
|
new_weights = copy.deepcopy(old_weights)
|
|
for layer_id, layer in enumerate(new_weights):
|
|
for cell_id, cell in enumerate(layer):
|
|
for weight_id, weight in enumerate(cell):
|
|
if prng() < 0.5:
|
|
new_weights[layer_id][cell_id][weight_id] = weight + prng() * e
|
|
else:
|
|
new_weights[layer_id][cell_id][weight_id] = weight - prng() * e
|
|
return new_weights
|
|
|
|
|
|
if __name__ == '__main__':
|
|
with Experiment('known-fixpoint-variation') as exp:
|
|
exp.depth = 10
|
|
exp.trials = 100
|
|
exp.max_steps = 100
|
|
exp.epsilon = 1e-4
|
|
exp.xs = []
|
|
exp.ys = []
|
|
exp.zs = []
|
|
exp.notable_nets = []
|
|
current_scale = 1.0
|
|
for _ in range(exp.depth):
|
|
print('variation scale ' + str(current_scale))
|
|
for _ in tqdm(range(exp.trials)):
|
|
net = generate_fixpoint_net().with_params(epsilon=exp.epsilon)
|
|
net = ParticleDecorator(net)
|
|
net.set_weights(vary(net.get_weights(), current_scale))
|
|
time_to_something = 0
|
|
time_as_fixpoint = 0
|
|
still_fixpoint = True
|
|
for _ in range(exp.max_steps):
|
|
net.self_attack()
|
|
if net.is_zero() or net.is_diverged():
|
|
break
|
|
if net.is_fixpoint():
|
|
if still_fixpoint:
|
|
time_as_fixpoint += 1
|
|
else:
|
|
print('remarkable')
|
|
exp.notable_nets += [net.get_weights()]
|
|
still_fixpoint = True
|
|
else:
|
|
still_fixpoint = False
|
|
time_to_something += 1
|
|
exp.xs += [current_scale]
|
|
# time steps taken to reach divergence or zero (reaching another fix-point is basically never happening)
|
|
exp.ys += [time_to_something]
|
|
# time steps still regarded as sthe initial fix-point
|
|
exp.zs += [time_as_fixpoint]
|
|
keras.backend.clear_session()
|
|
current_scale /= 10.0
|
|
for d in range(exp.depth):
|
|
exp.log('variation 10e-' + str(d))
|
|
exp.log('avg time to vergence ' + str(avg(exp.ys[d*exp.trials:(d+1) * exp.trials])))
|
|
exp.log('avg time as fixpoint ' + str(avg(exp.zs[d*exp.trials:(d+1) * exp.trials])))
|