julius/prototorch_models
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Automatic Formating.

Browse Source
This commit is contained in:
Alexander Engelsberger
2021-04-23 17:27:47 +02:00
parent db4499a103
commit c4c51a16fe
12 changed files with 404 additions and 159 deletions
Download Patch File Download Diff File Expand all files Collapse all files

62
examples/cbc_circle.py
View File

@@ -4,26 +4,24 @@ import numpy as np
import pytorch_lightning as pl
import torch
from matplotlib import pyplot as plt
from prototorch.models.cbc import CBC, rescaled_cosine_similarity, euclidean_similarity
from prototorch.models.glvq import GLVQ
from sklearn.datasets import make_circles
from torch.utils.data import DataLoader, TensorDataset
from torch.utils.data import DataLoader
class NumpyDataset(TensorDataset):
def __init__(self, *arrays):
# tensors = [torch.from_numpy(arr) for arr in arrays]
tensors = [torch.Tensor(arr) for arr in arrays]
super().__init__(*tensors)
from prototorch.datasets.abstract import NumpyDataset
from prototorch.models.callbacks.visualization import VisPointProtos
from prototorch.models.cbc import CBC, euclidean_similarity
from prototorch.models.glvq import GLVQ
class VisualizationCallback(pl.Callback):
def __init__(self,
x_train,
y_train,
prototype_model=True,
title="Prototype Visualization",
cmap="viridis"):
def __init__(
self,
x_train,
y_train,
prototype_model=True,
title="Prototype Visualization",
cmap="viridis",
):
super().__init__()
self.x_train = x_train
self.y_train = y_train
@@ -38,20 +36,22 @@ class VisualizationCallback(pl.Callback):
color = pl_module.prototype_labels
else:
protos = pl_module.components
color = 'k'
color = "k"
ax = self.fig.gca()
ax.cla()
ax.set_title(self.title)
ax.set_xlabel("Data dimension 1")
ax.set_ylabel("Data dimension 2")
ax.scatter(x_train[:, 0], x_train[:, 1], c=y_train, edgecolor="k")
ax.scatter(protos[:, 0],
protos[:, 1],
c=color,
cmap=self.cmap,
edgecolor="k",
marker="D",
s=50)
ax.scatter(
protos[:, 0],
protos[:, 1],
c=color,
cmap=self.cmap,
edgecolor="k",
marker="D",
s=50,
)
x = np.vstack((x_train, protos))
x_min, x_max = x[:, 0].min() - 1, x[:, 0].max() + 1
y_min, y_max = x[:, 1].min() - 1, x[:, 1].max() + 1
@@ -95,7 +95,7 @@ if __name__ == "__main__":
similarity=euclidean_similarity,
)
#model = GLVQ(hparams, data=[x_train, y_train])
model = GLVQ(hparams, data=[x_train, y_train])
# Fix the component locations
# model.proto_layer.requires_grad_(False)
@@ -107,13 +107,21 @@ if __name__ == "__main__":
print(model)
# Callbacks
vis = VisualizationCallback(x_train, y_train, prototype_model=False)
dvis = VisPointProtos(
data=(x_train, y_train),
save=True,
snap=False,
voronoi=True,
resolution=50,
pause_time=0.1,
make_gif=True,
)
# Setup trainer
trainer = pl.Trainer(
max_epochs=500,
max_epochs=10,
callbacks=[
vis,
dvis,
],
)

26
examples/cbc_iris.py
View File

@@ -4,16 +4,11 @@ import numpy as np
import pytorch_lightning as pl
import torch
from matplotlib import pyplot as plt
from prototorch.models.cbc import CBC
from sklearn.datasets import load_iris
from torch.utils.data import DataLoader, TensorDataset
from torch.utils.data import DataLoader
class NumpyDataset(TensorDataset):
def __init__(self, *arrays):
# tensors = [torch.from_numpy(arr) for arr in arrays]
tensors = [torch.Tensor(arr) for arr in arrays]
super().__init__(*tensors)
from prototorch.datasets.abstract import NumpyDataset
from prototorch.models.cbc import CBC
class VisualizationCallback(pl.Callback):
@@ -47,7 +42,8 @@ class VisualizationCallback(pl.Callback):
cmap=self.cmap,
edgecolor="k",
marker="D",
s=50)
s=50,
)
x = np.vstack((x_train, protos))
x_min, x_max = x[:, 0].min() - 1, x[:, 0].max() + 1
y_min, y_max = x[:, 1].min() - 1, x[:, 1].max() + 1
@@ -73,11 +69,13 @@ if __name__ == "__main__":
train_loader = DataLoader(train_ds, num_workers=0, batch_size=150)
# Hyperparameters
hparams = dict(input_dim=x_train.shape[1],
nclasses=3,
prototypes_per_class=3,
prototype_initializer="stratified_mean",
lr=0.01)
hparams = dict(
input_dim=x_train.shape[1],
nclasses=3,
prototypes_per_class=3,
prototype_initializer="stratified_mean",
lr=0.01,
)
# Initialize the model
model = CBC(hparams, data=[x_train, y_train])

12
examples/cbc_mnist.py
View File

@@ -7,12 +7,12 @@ import argparse
import pytorch_lightning as pl
import torchvision
from matplotlib import pyplot as plt
from prototorch.models.cbc import ImageCBC, euclidean_similarity, rescaled_cosine_similarity
from torch.utils.data import DataLoader
from torchvision import transforms
from torchvision.datasets import MNIST
from prototorch.models.cbc import CBC, ImageCBC, euclidean_similarity
class VisualizationCallback(pl.Callback):
def __init__(self, to_shape=(-1, 1, 28, 28), nrow=2):
@@ -89,8 +89,8 @@ if __name__ == "__main__":
)
# Dataloaders
train_loader = DataLoader(mnist_train, batch_size=1024)
test_loader = DataLoader(mnist_test, batch_size=1024)
train_loader = DataLoader(mnist_train, batch_size=32)
test_loader = DataLoader(mnist_test, batch_size=32)
# Grab the full dataset to warm-start prototypes
x, y = next(iter(DataLoader(mnist_train, batch_size=len(mnist_train))))
@@ -102,12 +102,12 @@ if __name__ == "__main__":
nclasses=10,
prototypes_per_class=args.ppc,
prototype_initializer="randn",
lr=1,
lr=0.01,
similarity=euclidean_similarity,
)
# Initialize the model
model = ImageCBC(hparams, data=[x, y])
model = CBC(hparams, data=[x, y])
# Model summary
print(model)

135
examples/cbc_spiral.py Normal file
View File

@@ -0,0 +1,135 @@
"""CBC example using the Iris dataset."""
import numpy as np
import pytorch_lightning as pl
import torch
from matplotlib import pyplot as plt
from torch.utils.data import DataLoader
from prototorch.datasets.abstract import NumpyDataset
from prototorch.models.cbc import CBC
class VisualizationCallback(pl.Callback):
def __init__(
self,
x_train,
y_train,
prototype_model=True,
title="Prototype Visualization",
cmap="viridis",
):
super().__init__()
self.x_train = x_train
self.y_train = y_train
self.title = title
self.fig = plt.figure(self.title)
self.cmap = cmap
self.prototype_model = prototype_model
def on_epoch_end(self, trainer, pl_module):
if self.prototype_model:
protos = pl_module.prototypes
color = pl_module.prototype_labels
else:
protos = pl_module.components
color = "k"
ax = self.fig.gca()
ax.cla()
ax.set_title(self.title)
ax.set_xlabel("Data dimension 1")
ax.set_ylabel("Data dimension 2")
ax.scatter(x_train[:, 0], x_train[:, 1], c=y_train, edgecolor="k")
ax.scatter(
protos[:, 0],
protos[:, 1],
c=color,
cmap=self.cmap,
edgecolor="k",
marker="D",
s=50,
)
x = np.vstack((x_train, protos))
x_min, x_max = x[:, 0].min() - 1, x[:, 0].max() + 1
y_min, y_max = x[:, 1].min() - 1, x[:, 1].max() + 1
xx, yy = np.meshgrid(np.arange(x_min, x_max, 1 / 50),
np.arange(y_min, y_max, 1 / 50))
mesh_input = np.c_[xx.ravel(), yy.ravel()]
y_pred = pl_module.predict(torch.Tensor(mesh_input))
y_pred = y_pred.reshape(xx.shape)
ax.contourf(xx, yy, y_pred, cmap=self.cmap, alpha=0.35)
ax.set_xlim(left=x_min + 0, right=x_max - 0)
ax.set_ylim(bottom=y_min + 0, top=y_max - 0)
plt.pause(0.1)
def make_spirals(n_samples=500, noise=0.3):
def get_samples(n, delta_t):
points = []
for i in range(n):
r = i / n_samples * 5
t = 1.75 * i / n * 2 * np.pi + delta_t
x = r * np.sin(t) + np.random.rand(1) * noise
y = r * np.cos(t) + np.random.rand(1) * noise
points.append([x, y])
return points
n = n_samples // 2
positive = get_samples(n=n, delta_t=0)
negative = get_samples(n=n, delta_t=np.pi)
x = np.concatenate(
[np.array(positive).reshape(n, -1),
np.array(negative).reshape(n, -1)],
axis=0)
y = np.concatenate([np.zeros(n), np.ones(n)])
return x, y
if __name__ == "__main__":
# Dataset
x_train, y_train = make_spirals(n_samples=1000, noise=0.3)
train_ds = NumpyDataset(x_train, y_train)
# Dataloaders
train_loader = DataLoader(train_ds, num_workers=0, batch_size=150)
# Hyperparameters
hparams = dict(
input_dim=x_train.shape[1],
nclasses=2,
prototypes_per_class=40,
prototype_initializer="stratified_random",
lr=0.05,
)
# Initialize the model
model_class = CBC
model = model_class(hparams, data=[x_train, y_train])
# Pure-positive reasonings
new_reasoning = torch.zeros_like(
model.reasoning_layer.reasoning_probabilities)
for i, label in enumerate(model.proto_layer.prototype_labels):
new_reasoning[0][0][i][int(label)] = 1.0
model.reasoning_layer.reasoning_probabilities.data = new_reasoning
# Model summary
print(model)
# Callbacks
vis = VisualizationCallback(x_train,
y_train,
prototype_model=hasattr(model, "prototypes"))
# Setup trainer
trainer = pl.Trainer(
max_epochs=500,
callbacks=[
vis,
],
)
# Training loop
trainer.fit(model, train_loader)

142
examples/cbc_spiral_with_GLVQ_start.py Normal file
View File

@@ -0,0 +1,142 @@
"""CBC example using the Iris dataset."""
import numpy as np
import pytorch_lightning as pl
import torch
from matplotlib import pyplot as plt
from torch.utils.data import DataLoader
from prototorch.datasets.abstract import NumpyDataset
from prototorch.models.cbc import CBC
from prototorch.models.glvq import GLVQ
class VisualizationCallback(pl.Callback):
def __init__(
self,
x_train,
y_train,
prototype_model=True,
title="Prototype Visualization",
cmap="viridis",
):
super().__init__()
self.x_train = x_train
self.y_train = y_train
self.title = title
self.fig = plt.figure(self.title)
self.cmap = cmap
self.prototype_model = prototype_model
def on_epoch_end(self, trainer, pl_module):
if self.prototype_model:
protos = pl_module.prototypes
color = pl_module.prototype_labels
else:
protos = pl_module.components
color = "k"
ax = self.fig.gca()
ax.cla()
ax.set_title(self.title)
ax.set_xlabel("Data dimension 1")
ax.set_ylabel("Data dimension 2")
ax.scatter(x_train[:, 0], x_train[:, 1], c=y_train, edgecolor="k")
ax.scatter(
protos[:, 0],
protos[:, 1],
c=color,
cmap=self.cmap,
edgecolor="k",
marker="D",
s=50,
)
x = np.vstack((x_train, protos))
x_min, x_max = x[:, 0].min() - 1, x[:, 0].max() + 1
y_min, y_max = x[:, 1].min() - 1, x[:, 1].max() + 1
xx, yy = np.meshgrid(np.arange(x_min, x_max, 1 / 50),
np.arange(y_min, y_max, 1 / 50))
mesh_input = np.c_[xx.ravel(), yy.ravel()]
y_pred = pl_module.predict(torch.Tensor(mesh_input))
y_pred = y_pred.reshape(xx.shape)
ax.contourf(xx, yy, y_pred, cmap=self.cmap, alpha=0.35)
ax.set_xlim(left=x_min + 0, right=x_max - 0)
ax.set_ylim(bottom=y_min + 0, top=y_max - 0)
plt.pause(0.1)
def make_spirals(n_samples=500, noise=0.3):
def get_samples(n, delta_t):
points = []
for i in range(n):
r = i / n_samples * 5
t = 1.75 * i / n * 2 * np.pi + delta_t
x = r * np.sin(t) + np.random.rand(1) * noise
y = r * np.cos(t) + np.random.rand(1) * noise
points.append([x, y])
return points
n = n_samples // 2
positive = get_samples(n=n, delta_t=0)
negative = get_samples(n=n, delta_t=np.pi)
x = np.concatenate(
[np.array(positive).reshape(n, -1),
np.array(negative).reshape(n, -1)],
axis=0)
y = np.concatenate([np.zeros(n), np.ones(n)])
return x, y
def train(model, x_train, y_train, train_loader, epochs=100):
# Callbacks
vis = VisualizationCallback(x_train,
y_train,
prototype_model=hasattr(model, "prototypes"))
# Setup trainer
trainer = pl.Trainer(
max_epochs=epochs,
callbacks=[
vis,
],
)
# Training loop
trainer.fit(model, train_loader)
if __name__ == "__main__":
# Dataset
x_train, y_train = make_spirals(n_samples=1000, noise=0.3)
train_ds = NumpyDataset(x_train, y_train)
# Dataloaders
train_loader = DataLoader(train_ds, num_workers=0, batch_size=150)
# Hyperparameters
hparams = dict(
input_dim=x_train.shape[1],
nclasses=2,
prototypes_per_class=40,
prototype_initializer="stratified_random",
lr=0.05,
)
# Initialize the model
glvq_model = GLVQ(hparams, data=[x_train, y_train])
cbc_model = CBC(hparams, data=[x_train, y_train])
# Train GLVQ
train(glvq_model, x_train, y_train, train_loader, epochs=10)
# Transfer Prototypes
cbc_model.proto_layer.load_state_dict(glvq_model.proto_layer.state_dict())
# Pure-positive reasonings
new_reasoning = torch.zeros_like(
cbc_model.reasoning_layer.reasoning_probabilities)
for i, label in enumerate(cbc_model.proto_layer.prototype_labels):
new_reasoning[0][0][i][int(label)] = 1.0
new_reasoning[1][0][i][1 - int(label)] = 1.0
cbc_model.reasoning_layer.reasoning_probabilities.data = new_reasoning
# Train CBC
train(cbc_model, x_train, y_train, train_loader, epochs=50)

30
examples/glvq_iris.py
View File

@@ -6,15 +6,11 @@ import numpy as np
import pytorch_lightning as pl
import torch
from matplotlib import pyplot as plt
from prototorch.models.glvq import GLVQ
from sklearn.datasets import load_iris
from torch.utils.data import DataLoader, TensorDataset
from torch.utils.data import DataLoader
class NumpyDataset(TensorDataset):
def __init__(self, *arrays):
tensors = [torch.from_numpy(arr) for arr in arrays]
super().__init__(*tensors)
from prototorch.datasets.abstract import NumpyDataset
from prototorch.models.glvq import GLVQ
class GLVQIris(GLVQ):
@@ -56,13 +52,15 @@ class VisualizationCallback(pl.Callback):
ax.set_xlabel("Data dimension 1")
ax.set_ylabel("Data dimension 2")
ax.scatter(x_train[:, 0], x_train[:, 1], c=y_train, edgecolor="k")
ax.scatter(protos[:, 0],
protos[:, 1],
c=plabels,
cmap=self.cmap,
edgecolor="k",
marker="D",
s=50)
ax.scatter(
protos[:, 0],
protos[:, 1],
c=plabels,
cmap=self.cmap,
edgecolor="k",
marker="D",
s=50,
)
x = np.vstack((x_train, protos))
x_min, x_max = x[:, 0].min() - 1, x[:, 0].max() + 1
y_min, y_max = x[:, 1].min() - 1, x[:, 1].max() + 1
@@ -105,8 +103,8 @@ if __name__ == "__main__":
parser,
max_epochs=10,
callbacks=[
vis, # comment this line out to disable the visualization
],
vis,
], # comment this line out to disable the visualization
)
# trainer.tune(model)

38
examples/glvq_iris_v1.py
View File

@@ -4,15 +4,11 @@ import numpy as np
import pytorch_lightning as pl
import torch
from matplotlib import pyplot as plt
from prototorch.models.glvq import GLVQ
from sklearn.datasets import load_iris
from torch.utils.data import DataLoader, TensorDataset
from torch.utils.data import DataLoader
class NumpyDataset(TensorDataset):
def __init__(self, *arrays):
tensors = [torch.from_numpy(arr) for arr in arrays]
super().__init__(*tensors)
from prototorch.datasets.abstract import NumpyDataset
from prototorch.models.glvq import GLVQ
class VisualizationCallback(pl.Callback):
@@ -37,13 +33,15 @@ class VisualizationCallback(pl.Callback):
ax.set_xlabel("Data dimension 1")
ax.set_ylabel("Data dimension 2")
ax.scatter(x_train[:, 0], x_train[:, 1], c=y_train, edgecolor="k")
ax.scatter(protos[:, 0],
protos[:, 1],
c=plabels,
cmap=self.cmap,
edgecolor="k",
marker="D",
s=50)
ax.scatter(
protos[:, 0],
protos[:, 1],
c=plabels,
cmap=self.cmap,
edgecolor="k",
marker="D",
s=50,
)
x = np.vstack((x_train, protos))
x_min, x_max = x[:, 0].min() - 1, x[:, 0].max() + 1
y_min, y_max = x[:, 1].min() - 1, x[:, 1].max() + 1
@@ -69,11 +67,13 @@ if __name__ == "__main__":
train_loader = DataLoader(train_ds, num_workers=0, batch_size=150)
# Hyperparameters
hparams = dict(input_dim=x_train.shape[1],
nclasses=3,
prototypes_per_class=3,
prototype_initializer="stratified_mean",
lr=0.1)
hparams = dict(
input_dim=x_train.shape[1],
nclasses=3,
prototypes_per_class=3,
prototype_initializer="stratified_mean",
lr=0.1,
)
# Initialize the model
model = GLVQ(hparams, data=[x_train, y_train])

27
examples/glvq_mnist.py
View File

@@ -11,13 +11,12 @@ import argparse
import pytorch_lightning as pl
import torchvision
from matplotlib import pyplot as plt
from prototorch.functions.initializers import stratified_mean
from prototorch.models.glvq import ImageGLVQ
from torch.utils.data import DataLoader
from torchvision import transforms
from torchvision.datasets import MNIST
from prototorch.models.glvq import ImageGLVQ
class VisualizationCallback(pl.Callback):
def __init__(self, to_shape=(-1, 1, 28, 28), nrow=2):
@@ -31,10 +30,12 @@ class VisualizationCallback(pl.Callback):
grid = torchvision.utils.make_grid(protos_img, nrow=self.nrow)
# grid = grid.permute((1, 2, 0))
tb = pl_module.logger.experiment
tb.add_image(tag="MNIST Prototypes",
img_tensor=grid,
global_step=trainer.current_epoch,
dataformats="CHW")
tb.add_image(
tag="MNIST Prototypes",
img_tensor=grid,
global_step=trainer.current_epoch,
dataformats="CHW",
)
if __name__ == "__main__":
@@ -91,11 +92,13 @@ if __name__ == "__main__":
x = x.view(len(mnist_train), -1)
# Initialize the model
model = ImageGLVQ(input_dim=28 * 28,
nclasses=10,
prototypes_per_class=args.ppc,
prototype_initializer="stratified_mean",
data=[x, y])
model = ImageGLVQ(
input_dim=28 * 28,
nclasses=10,
prototypes_per_class=args.ppc,
prototype_initializer="stratified_mean",
data=[x, y],
)
# Model summary
print(model)