def test_reset(self):
# Node and Marble have some arbitrary nonzero initial motion vars.
node = Node(torch.tensor([1.]), torch.tensor([2.]), torch.tensor([3.]),
4, NewtonianGravity(), 1, 1, 1, 1)
marble = Marble(torch.tensor([1.1]), torch.tensor([2.2]),
torch.tensor([3.3]), 4.4, NewtonianGravity(), None)
model = NenwinModel([node], [marble])
model.make_timestep(10)
model.make_timestep(10)
model.make_timestep(10)
# Verify that the motion variables have changed
self.assertFalse(check_close(node.pos, node.init_pos))
self.assertFalse(check_close(marble.vel, marble.init_vel))
self.assertFalse(check_close(marble.acc, marble.init_acc))
model.reset()
# Now they should be the original values again.
self.assertTrue(check_close(node.pos, node.init_pos))
self.assertTrue(check_close(node.vel, node.init_vel))
self.assertTrue(check_close(node.acc, node.init_acc))
self.assertTrue(check_close(marble.pos, marble.init_pos))
self.assertTrue(check_close(marble.vel, marble.init_vel))
self.assertTrue(check_close(marble.acc, marble.init_acc))
def test_eaten_marbles_disappear(self):
# Both are generated at the smame pos, so immediately eaten.
node = MarbleEaterNode(ZERO, ZERO, ZERO, 10, NewtonianGravity(), 1, 1,
1, 0, 10)
marble = Marble(ZERO, ZERO, ZERO, 10, NewtonianGravity(), datum=None)
model = NenwinModel([node], [marble])
model.make_timestep(1.0)
self.assertSetEqual(model.marbles, set([]))
self.assertNotIn(marble, model._NenwinModel__all_particles)
def gen_architecture_c(
input_placer_type: type
) -> Tuple[NenwinModel, VelInputPlacer, Tuple[Node]]:
"""
Generate the following architecture:
* The input region is at (0, 0) and has size (6, 1)
(So it has vertices {(0, 0), (0, 1), (6, 0), (6, 1)})
* There are two MarbleEaterNodes, at (1, 4) and (5, 4)
* There are five normal Nodes,
at (1, 2), (2, 3), (3, 2), (4, 3) and (5, 2).
Returns:
* Model holding the architecture descibed above.
* VelInputPlacer with the input region as described above.
* Tuple of the two MarbleEaterNodes.
"""
eater_positions = [(1, 4), (5, 4)]
node_positions = [(1, 2), (2, 3), (3, 2), (4, 3), (5, 2)]
input_region_pos = np.array((0, 0))
input_region_size = np.array((6, 1))
mass = 1
radius = 0.5
attraction_function = NewtonianGravity()
nodes = gen_nodes(attraction_function, mass, node_positions)
eater_nodes = gen_eater_nodes(attraction_function, mass, radius,
eater_positions)
model = NenwinModel(nodes + eater_nodes)
input_placer = input_placer_type(input_region_pos, input_region_size)
return model, input_placer, eater_nodes
def visualization_demo():
attract_funct = NewtonianGravity()
marbles = set()
marble_1 = Marble(np.array([460, 460]),
np.array([0, 0]),
np.array([0, 0]),
mass=20,
attraction_function=attract_funct,
datum=None,
marble_stiffness=1,
node_stiffness=0,
marble_attraction=0,
node_attraction=1)
marble_2 = Marble(np.array([550, 550]),
np.array([10, 10]),
np.array([0, 0]),
mass=40,
attraction_function=attract_funct,
datum=None,
marble_stiffness=1,
node_stiffness=0,
marble_attraction=0,
node_attraction=1)
marble_3 = Marble(np.array([450, 500]),
np.array([2, -2]),
np.array([0, 0]),
mass=40,
attraction_function=attract_funct,
datum=None,
marble_stiffness=1,
node_stiffness=0,
marble_attraction=0,
node_attraction=1)
marbles = set((marble_1, marble_2, marble_3))
node_1 = Node(np.array([500, 500]),
np.array([0, 0]),
np.array([0, 0]),
mass=200,
attraction_function=attract_funct,
marble_stiffness=1,
node_stiffness=1,
marble_attraction=1,
node_attraction=1)
node_2 = Node(np.array([400, 400]),
np.array([0, 0]),
np.array([0, 0]),
mass=200,
attraction_function=attract_funct,
marble_stiffness=0.7,
node_stiffness=1,
marble_attraction=1,
node_attraction=1)
nodes = set((node_1, node_2))
model = NenwinModel(nodes, marbles)
simulation = Simulation(model, None, None, MockPipe())
visualization = NenwinVisualization((1500, 1000), simulation, model)
visualization.run(10, 0.01)
def test_particle_gradients_moving_node(self):
"""
Base case: Marble being attracted by a moving Node,
when using Marble's variables to compute loss,
also Node.pos should receive gradients
when computing backprop on the loss.
"""
self.node = Node(pos=ZERO,
vel=torch.tensor([0.1]),
acc=ZERO,
mass=1,
attraction_function=NewtonianGravity(),
marble_stiffness=1,
node_stiffness=1,
marble_attraction=1,
node_attraction=0)
self.model = NenwinModel([self.node], [self.marble])
self.model.make_timestep(1.0)
self.model.make_timestep(1.0)
self.model.make_timestep(1.0)
loss = 2 * self.marble.acc
loss.backward()
self.assertIsNotNone(self.marble.init_pos.grad)
self.assertIsNotNone(self.node.init_pos.grad)
self.assertIsNotNone(self.node.init_vel.grad)
self.assertIsNotNone(self.node._PhysicalParticle__mass.grad)
def setup_full_emitter(
self) -> Tuple[Emitter, Marble, float, float, float]:
"""
Create an instance of a MarbleEmitter with a full prototype Marble
(not a mock Marble). Also returns the delay, stored_mass
and initial_time_passed,
which have arbitrary nonzero floating-point values.
Returns:
* A MarbleEmitter instance
* A Marble, the prototype associated with the emitter
* The delay of the emitter
* The initally stored mass of the emitter
* The initial time passed of the emitter
"""
pos = torch.tensor([0], dtype=torch.float)
vel = torch.tensor([1], dtype=torch.float)
acc = torch.tensor([2], dtype=torch.float)
mass = 3.0
attraction_function = NewtonianGravity()
marble_stiffness = 0.4
node_stiffness = 0.5
marble_attraction = 0.6
node_attraction = 0.7
prototype = Marble(pos, vel, acc, mass, attraction_function, None,
marble_stiffness, node_stiffness, marble_attraction,
node_attraction)
delay = 8.1
stored_mass = 9.1
initial_time_passed = 11.1
emitter = MarbleEmitter(prototype, delay, stored_mass,
initial_time_passed)
return emitter, prototype, delay, stored_mass, initial_time_passed
def test_repr(self):
pos = torch.tensor([0], dtype=torch.float)
vel = torch.tensor([1], dtype=torch.float)
acc = torch.tensor([2], dtype=torch.float)
mass = 3.0
attraction_function = NewtonianGravity()
marble_stiffness = 0.4
node_stiffness = 0.5
marble_attraction = 0.6
node_attraction = 0.7
node = Node(pos, vel, acc, mass, attraction_function, marble_stiffness,
node_stiffness, marble_attraction, node_attraction)
# Some numerical errors occurs when converting from float to FloatTensor
marble_stiffness_float_repr = \
convert_scalar_param_to_repr(marble_stiffness)
node_stiffness_float_repr = convert_scalar_param_to_repr(
node_stiffness)
marble_attraction_float_repr = \
convert_scalar_param_to_repr(marble_attraction)
node_attraction_float_repr = \
convert_scalar_param_to_repr(node_attraction)
expected = f"Node({repr(pos)},{repr(vel)},"\
+ f"{repr(acc)},{mass},NewtonianGravity(),"\
+ f"{marble_stiffness_float_repr}," \
+ f"{node_stiffness_float_repr},{marble_attraction_float_repr}," \
+ f"{node_attraction_float_repr})"
result = repr(node)
self.assertEqual(expected, result)
def gen_architecture_b(
input_placer_type: type
) -> Tuple[NenwinModel, VelInputPlacer, Tuple[Node]]:
"""
Same as architecture A, but with four additional Nodes
(normal Nodes, no MarbleEaterNodes) at:
* (2.5, 2.5), (-2.5, 2.5), (2.5, -2.5), (-2.5, -2.5)
So the normal Nodes surround the input-region,
and the MarbleEaterNodes are to the left and right of this.
Returns:
* Model holding the architecture descibed above.
* VelInputPlacer with the input region as described above.
* Tuple of the two MarbleEaterNodes.
"""
eater_positions = [(-10, 0), (10, 0)]
node_positions = [(0, -5), (-5, 0), (5, 0), (0, 5), (2.5, 2.5),
(-2.5, 2.5), (2.5, -2.5), (-2.5, -2.5)]
input_region_pos = np.array((-2.5, -1))
input_region_size = np.array((5, 2))
mass = 1
radius = 0.5
attraction_function = NewtonianGravity()
nodes = gen_nodes(attraction_function, mass, node_positions)
eater_nodes = gen_eater_nodes(attraction_function, mass, radius,
eater_positions)
model = NenwinModel(nodes + eater_nodes)
input_placer = input_placer_type(input_region_pos, input_region_size)
return model, input_placer, eater_nodes
def gen_node_at(pos: torch.Tensor, mass: float = 10) -> MarbleEaterNode:
output = MarbleEaterNode(pos, vel=ZERO, acc=ZERO,
mass=mass,
attraction_function=NewtonianGravity(),
marble_attraction=1,
marble_stiffness=1,
node_attraction=0,
node_stiffness=0,
radius=1)
return output
def test_repr(self):
pos = torch.tensor([0], dtype=torch.float)
vel = torch.tensor([1], dtype=torch.float)
acc = torch.tensor([2], dtype=torch.float)
mass = 3.0
attraction_funct = NewtonianGravity()
expected = f"PhysicalParticle({repr(pos)},{repr(vel)},"\
+ f"{repr(acc)},{mass},NewtonianGravity(),{repr(DEVICE)})"
result = repr(PhysicalParticle(pos, vel, acc, mass, attraction_funct))
self.assertEqual(expected, result)
def gen_marble_at(pos: torch.Tensor,
vel: torch.Tensor = ZERO,
mass: float = 10,
datum: Any = None) -> Marble:
output = Marble(pos=pos, vel=vel, acc=ZERO,
mass=mass,
attraction_function=NewtonianGravity(),
marble_attraction=0,
marble_stiffness=1,
node_attraction=0,
node_stiffness=0,
datum=datum)
return output
def setUp(self):
self.node = Node(pos=ZERO,
vel=ZERO,
acc=ZERO,
mass=1,
attraction_function=NewtonianGravity(),
marble_stiffness=1,
node_stiffness=1,
marble_attraction=1,
node_attraction=0)
self.marble = Marble(pos=np.array([5]),
vel=ZERO,
acc=ZERO,
mass=1,
attraction_function=ATTRACT_FUNCT,
datum=None)
self.model = NenwinModel([self.node], [self.marble])
def test_repr(self):
pos = torch.tensor([0], dtype=torch.float)
vel = torch.tensor([1], dtype=torch.float)
acc = torch.tensor([2], dtype=torch.float)
mass = 3.0
attraction_function = NewtonianGravity()
marble_stiffness = 0.4
node_stiffness = 0.5
marble_attraction = 0.6
node_attraction = 0.7
radius = 8.0
# Some numerical errors occurs when converting from float to FloatTensor
marble_stiffness_float_repr = \
convert_scalar_param_to_repr(marble_stiffness)
node_stiffness_float_repr = convert_scalar_param_to_repr(
node_stiffness)
marble_attraction_float_repr = \
convert_scalar_param_to_repr(marble_attraction)
node_attraction_float_repr = \
convert_scalar_param_to_repr(node_attraction)
emitter = MockEmitter(MockPrototype(), 0)
expected = f"MarbleEmitterNode({repr(pos)},{repr(vel)},"\
+ f"{repr(acc)},{mass},NewtonianGravity(),"\
+ f"{marble_stiffness_float_repr}," \
+ f"{node_stiffness_float_repr},{marble_attraction_float_repr}," \
+ f"{node_attraction_float_repr},"\
+ f"{radius}," \
+ f"{repr(emitter)})"
eater = MarbleEmitterNode(pos, vel, acc, mass, attraction_function,
marble_stiffness, node_stiffness,
marble_attraction, node_attraction, radius,
emitter)
result = repr(eater)
self.assertEqual(expected, result)
def test_reset_removed_added_marbles(self):
"""
NenwinModel.reset() should remove all Marbles
added with NenwinModel.add_marbles()
"""
original_marble = Marble(ZERO,
ZERO,
ZERO,
10,
NewtonianGravity(),
datum=None)
added_marble = original_marble.copy()
assert added_marble is not original_marble
model = NenwinModel([], [original_marble])
model.add_marbles([added_marble])
model.reset()
self.assertIn(original_marble, model.marbles)
self.assertIn(original_marble, model._NenwinModel__all_particles)
self.assertNotIn(added_marble, model.marbles)
self.assertNotIn(added_marble, model._NenwinModel__all_particles)
def gen_architecture_a(
input_placer_type: type
) -> Tuple[NenwinModel, VelInputPlacer, Tuple[Node]]:
"""
Generate the following architecture:
* The input region is at (-2.5, -1) and has size (5, 2)
(So it has vertices {(-2.5, -1), (-2.5, 1), (2.5, -1), (2.5, 1)})
* There are two MarbleEaterNodes, at (-10, 0) and (10, 0)
* There are four normal Nodes, at (0, -5), (-5, 0), (5, 0) and (0, 5).
So the normal Nodes surround the input-region,
and the MarbleEaterNodes are to the left and right of this.
Returns:
* Model holding the architecture descibed above.
* VelInputPlacer with the input region as described above.
* Tuple of the two MarbleEaterNodes.
"""
eater_positions = [(-10, 0), (10, 0)]
node_positions = [(0, -5), (-5, 0), (5, 0), (0, 5)]
input_region_pos = np.array((-2.5, -1))
input_region_size = np.array((5, 2))
mass = 1
radius = 0.5
attraction_function = NewtonianGravity()
nodes = gen_nodes(attraction_function, mass, node_positions)
eater_nodes = gen_eater_nodes(attraction_function, mass, radius,
eater_positions)
model = NenwinModel(nodes + eater_nodes)
input_placer = input_placer_type(input_region_pos, input_region_size)
return model, input_placer, eater_nodes
Author: Lulof Pirée
October 2020
An experiment to see if an architecture can be made that accelerates a
ḿarble in a straight line.
"""
import numpy as np
from typing import List
from nenwin.marble_eater_node import MarbleEaterNode
from nenwin.node import Node, Marble
from nenwin.attraction_functions.attraction_functions import NewtonianGravity
from nenwin.architectures.run_and_visualize import run
from nenwin.auxliary import generate_stiffness_dict
ZERO = np.array([0, 0])
ATTRACTION_FUNCTION = NewtonianGravity()
NODE_STIFFNESSES = generate_stiffness_dict(marble_stiffness=1,
node_stiffness=1,
marble_attraction=1,
node_attraction=0)
MARBLE_STIFFNESS = generate_stiffness_dict(marble_stiffness=0,
node_stiffness=0,
marble_attraction=0,
node_attraction=0)
def cannon_experiment():
"""
Simulate a NAND gate using a Nenwin architecture.
Also visualizes the simulation and prints the result.
"""
def setUp(self):
self.newtonian = NewtonianGravity()
def test_repr(self):
self.assertEqual("NewtonianGravity()", repr(NewtonianGravity()))
def setUp(self):
self.attract_funct = NewtonianGravity()
