def travelling_salesman():
N = 50
random = Random()
random.setSeed(0)
points = [[0 for x in xrange(2)] for x in xrange(N)]
for i in range(0, len(points)):
points[i][0] = random.nextDouble()
points[i][1] = random.nextDouble()
fill = [N] * N
ranges = array('i', fill)
odd_mimic = DiscreteUniformDistribution(ranges)
odd = DiscretePermutationDistribution(N)
ef = TravelingSalesmanRouteEvaluationFunction(points)
nf = SwapNeighbor()
mf = SwapMutation()
cf = TravelingSalesmanCrossOver(ef)
rhc_generic("TSPrhc50", ef, odd, nf, 1.0, 10000, 10, 5)
sa_generic("TSPsa50", ef, odd, nf, 1.0, 10000, 10, 5,
([1E12, 1E6], [0.999, 0.99, 0.95]))
ga_generic("TSPga50", ef, odd, mf, cf, 50.0, 10000, 10, 1,
([2000, 200], [0.5, 0.25], [0.25, 0.1, 0.02]))
mimic_discrete("TSPmimic50", ef, odd_mimic, ranges, 300.0, 10000, 10, 1,
([200], [100], [0.1, 0.5, 0.9]))
print "TSP all done"
class VectorUHash: def __init__(self, seed): self.random = Random(seed) self.seed = seed def __hash(self, b): self.random.setSeed(self.seed) p = 0 sum = 0 while p < len(b): c = 0 d = [0,0,0,0] while p < len(b) and c < 4: d[c] = ord(b[p]) c += 1 p += 1 x1 = (int(d[0]) & 0xff) + ((int(d[1]) & 0xff) << 8) +\ ((int(d[2]) & 0xff) << 16) + ((int(d[3]) & 0xff) << 24) c = 0 e = [0,0,0,0] while p < len(b) and c < 4: e[c] = ord(b[p]) c += 1 p += 1 x2 = (int(e[0]) & 0xff) + ((int(e[1]) & 0xff) << 8) +\ ((int(e[2]) & 0xff) << 16) + ((int(e[3]) & 0xff) << 24) a1 = self.random.nextLong() a2 = self.random.nextLong() a1 |= 1 a2 |= 1 x1 += a1 x2 += a2 if x2 != 0: x1 *= x2 sum += x1 return int(sum & 0x7fffffff) def hashbytes(self, bytes): return self.__hash(bytes)
def knapsack():
# Random number generator */
random = Random()
random.setSeed(0)
NUM_ITEMS = 40 # The number of items
COPIES_EACH = 4 # The number of copies each
MAX_WEIGHT = 50 # The maximum weight for a single element
MAX_VOLUME = 50 # The maximum volume for a single element
KNAPSACK_VOLUME = MAX_VOLUME * NUM_ITEMS * COPIES_EACH * .4 # The volume of the knapsack
# create copies
fill = [COPIES_EACH] * NUM_ITEMS
copies = array('i', fill)
# create weights and volumes
fill = [0] * NUM_ITEMS
weights = array('d', fill)
volumes = array('d', fill)
for i in range(0, NUM_ITEMS):
weights[i] = random.nextDouble() * MAX_WEIGHT
volumes[i] = random.nextDouble() * MAX_VOLUME
# create range
fill = [COPIES_EACH + 1] * NUM_ITEMS
ranges = array('i', fill)
ef = KnapsackEvaluationFunction(weights, volumes, KNAPSACK_VOLUME, copies)
odd = DiscreteUniformDistribution(ranges)
nf = DiscreteChangeOneNeighbor(ranges)
mf = DiscreteChangeOneMutation(ranges)
cf = UniformCrossOver()
rhc_generic("KnSrhc50", ef, odd, nf, 1.0, 10000, 10, 5)
sa_generic("KnSsa50", ef, odd, nf, 1.0, 10000, 10, 5,
([1E12, 1E6], [0.999, 0.99, 0.95]))
ga_generic("KnSga50", ef, odd, mf, cf, 50.0, 10000, 10, 1,
([2000, 200], [0.5, 0.25], [0.25, 0.1, 0.02]))
mimic_discrete("KnSmimic50", ef, odd, ranges, 300.0, 10000, 10, 1,
([200], [100], [0.1, 0.5, 0.9]))
print "KnS all done"
# The maximum weight for a single element
MAX_WEIGHT = 50
# The maximum volume for a single element
MAX_VOLUME = 50
# The volume of the knapsack
KNAPSACK_VOLUME = MAX_VOLUME * NUM_ITEMS * COPIES_EACH * .4
# create copies
fill = [COPIES_EACH] * NUM_ITEMS
copies = array('i', fill)
# create weights and volumes
fill = [0] * NUM_ITEMS
weights = array('d', fill)
volumes = array('d', fill)
random.setSeed(1) # !!!!!
for i in range(0, NUM_ITEMS):
weights[i] = random.nextDouble() * MAX_WEIGHT
volumes[i] = random.nextDouble() * MAX_VOLUME
# create range
fill = [COPIES_EACH + 1] * NUM_ITEMS
ranges = array('i', fill)
ef = KnapsackEvaluationFunction(
weights, volumes, KNAPSACK_VOLUME,
copies) # value, volume, max_volume, the number of copies per element
odd = DiscreteUniformDistribution(ranges)
nf = DiscreteChangeOneNeighbor(ranges)
mf = DiscreteChangeOneMutation(ranges)
cf = UniformCrossOver()
df = DiscreteDependencyTree(.1, ranges)
from common import mock_feature
from trustyai.explainers import LimeConfig, LimeExplainer
from trustyai.utils import TestUtils
from trustyai.model import (
PerturbationContext,
PredictionInput,
FeatureFactory,
SimplePrediction,
)
from java.util import Random
from org.kie.kogito.explainability.local import (
LocalExplanationException, )
jrandom = Random()
jrandom.setSeed(0)
def test_empty_prediction():
"""Check if the explanation returned is not null"""
config = (LimeConfig().withPerturbationContext(
PerturbationContext(jrandom,
DEFAULT_NO_OF_PERTURBATIONS)).withSamples(10))
lime_explainer = LimeExplainer(config)
input_ = PredictionInput([])
model = TestUtils.getSumSkipModel(0)
output = model.predictAsync([input_]).get().get(0)
prediction = SimplePrediction(input_, output)
with pytest.raises(LocalExplanationException):
lime_explainer.explain(prediction, model)
