build: bump version 1.0.0a7 → 1.0.0a8

omega trace normalization and matrix profile visualization

.bumpversion.cfg

@@ -1,13 +1,15 @@
 [bumpversion]
-current_version = 0.7.1
+current_version = 1.0.0a8
 commit = True
 tag = True
-parse = (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)
+parse = (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)((?P<release>[a-zA-Z0-9_.-]+))?
-serialize = {major}.{minor}.{patch}
+serialize = 
+	{major}.{minor}.{patch}-{release}
+	{major}.{minor}.{patch}
 message = build: bump version {current_version} → {new_version}
 
-[bumpversion:file:pyproject.toml]
+[bumpversion:file:setup.py]
 
-[bumpversion:file:./src/prototorch/models/__init__.py]
+[bumpversion:file:./prototorch/models/__init__.py]
 
 [bumpversion:file:./docs/source/conf.py]

.github/workflows/examples.yml

@@ -6,16 +6,16 @@ name: examples
 on:
   push:
     paths:
-      - "examples/**.py"
+      - 'examples/**.py'
 jobs:
   cpu:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+    - uses: actions/checkout@v2
-      - name: Set up Python 3.11
+    - name: Set up Python 3.10
-        uses: actions/setup-python@v4
+      uses: actions/setup-python@v2
       with:
-          python-version: "3.11"
+        python-version: "3.10"
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip

.github/workflows/pythonapp.yml

@@ -12,11 +12,11 @@ jobs:
   style:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+    - uses: actions/checkout@v2
-      - name: Set up Python 3.11
+    - name: Set up Python 3.10
-        uses: actions/setup-python@v4
+      uses: actions/setup-python@v2
       with:
-          python-version: "3.11"
+        python-version: "3.10"
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip
@@ -27,21 +27,22 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.8", "3.9", "3.10", "3.11"]
+        python-version: ["3.7", "3.8", "3.9", "3.10"]
         os: [ubuntu-latest, windows-latest]
         exclude:
+        - os: windows-latest
+          python-version: "3.7"
         - os: windows-latest
           python-version: "3.8"
         - os: windows-latest
           python-version: "3.9"
         - os: windows-latest
-            python-version: "3.10"
+          python-version: "3.11"
-
     runs-on: ${{ matrix.os }}
     steps:
     - uses: actions/checkout@v2
     - name: Set up Python ${{ matrix.python-version }}
-        uses: actions/setup-python@v4
+      uses: actions/setup-python@v2
       with:
         python-version: ${{ matrix.python-version }}
     - name: Install dependencies
@@ -56,18 +57,18 @@ jobs:
     needs: compatibility
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+    - uses: actions/checkout@v2
-      - name: Set up Python 3.11
+    - name: Set up Python 3.10
-        uses: actions/setup-python@v4
+      uses: actions/setup-python@v2
       with:
-          python-version: "3.11"
+        python-version: "3.10"
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip
         pip install .[all]
-          pip install build
+        pip install wheel
     - name: Build package
-        run: python -m build . -C verbose
+      run: python setup.py sdist bdist_wheel
     - name: Publish a Python distribution to PyPI
       uses: pypa/gh-action-pypi-publish@release/v1
       with:

.pre-commit-config.yaml

@@ -3,9 +3,10 @@
 
 repos:
 - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
+  rev: v4.3.0
   hooks:
   - id: trailing-whitespace
+    exclude: (^\.bumpversion\.cfg$|cli_messages\.py)
   - id: end-of-file-fixer
   - id: check-yaml
   - id: check-added-large-files
@@ -13,17 +14,17 @@ repos:
   - id: check-case-conflict
 
 - repo: https://github.com/myint/autoflake
-    rev: v2.1.1
+  rev: v1.7.7
   hooks:
   - id: autoflake
 
 - repo: http://github.com/PyCQA/isort
-    rev: 5.12.0
+  rev: 5.10.1
   hooks:
   - id: isort
 
 - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.3.0
+  rev: v0.982
   hooks:
   - id: mypy
     files: prototorch
@@ -33,17 +34,16 @@ repos:
   rev: v0.32.0
   hooks:
   - id: yapf
-        additional_dependencies: ["toml"]
 
 - repo: https://github.com/pre-commit/pygrep-hooks
-    rev: v1.10.0
+  rev: v1.9.0
   hooks:
   - id: python-use-type-annotations
   - id: python-no-log-warn
   - id: python-check-blanket-noqa
 
 - repo: https://github.com/asottile/pyupgrade
-    rev: v3.7.0
+  rev: v3.1.0
   hooks:
   - id: pyupgrade
 
@@ -52,3 +52,8 @@ repos:
   hooks:
   - id: gitlint
     args: [--contrib=CT1, --ignore=B6, --msg-filename]
+
+- repo: https://github.com/dosisod/refurb
+  rev: v1.4.0
+  hooks:
+    - id: refurb

docs/source/conf.py

@@ -23,7 +23,7 @@ author = "Jensun Ravichandran"
 
 # The full version, including alpha/beta/rc tags
 #
-release = "0.7.1"
+release = "1.0.0-a8"
 
 # -- General configuration ---------------------------------------------------
 

docs/source/index.rst

@@ -23,6 +23,13 @@ ProtoTorch Models Plugins
 
    custom
 
+.. toctree::
+   :hidden:
+   :maxdepth: 3
+   :caption: Proto Y Architecture
+
+   y-architecture
+
 About
 -----------------------------------------
 `Prototorch Models <https://github.com/si-cim/prototorch_models>`_ is a Plugin
@@ -33,8 +40,10 @@ prototype-based Machine Learning algorithms using `PyTorch-Lightning
 Library
 -----------------------------------------
 Prototorch Models delivers many application ready models.
-These models have been published in the past and have been adapted to the Prototorch library.
+These models have been published in the past and have been adapted to the
+Prototorch library.
 
 Customizable
 -----------------------------------------
-Prototorch Models also contains the building blocks to build own models with PyTorch-Lightning and Prototorch.
+Prototorch Models also contains the building blocks to build own models with
+PyTorch-Lightning and Prototorch.

docs/source/library.rst

@@ -71,7 +71,7 @@ Probabilistic Models
 Probabilistic variants assume, that the prototypes generate a probability distribution over the classes.
 For a test sample they return a distribution instead of a class assignment.
 
-The following two algorihms were presented by :cite:t:`seo2003` .
+The following two algorithms were presented by :cite:t:`seo2003` .
 Every prototypes is a center of a gaussian distribution of its class, generating a mixture model.
 
 .. autoclass:: prototorch.models.probabilistic.SLVQ
@@ -80,7 +80,7 @@ Every prototypes is a center of a gaussian distribution of its class, generating
 .. autoclass:: prototorch.models.probabilistic.RSLVQ
    :members:
 
-:cite:t:`villmann2018` proposed two changes to RSLVQ: First incooperate the winning rank into the prior probability calculation.
+:cite:t:`villmann2018` proposed two changes to RSLVQ: First incorporate the winning rank into the prior probability calculation.
 And second use divergence as loss function.
 
 .. autoclass:: prototorch.models.probabilistic.PLVQ
@@ -106,7 +106,7 @@ Visualization
 Visualization is very specific to its application.
 PrototorchModels delivers visualization for two dimensional data and image data.
 
-The visulizations can be shown in a seperate window and inside a tensorboard.
+The visualizations can be shown in a separate window and inside a tensorboard.
 
 .. automodule:: prototorch.models.vis
    :members:

docs/source/y-architecture.rst

@@ -0,0 +1,71 @@
+.. Documentation of the updated Architecture.
+
+Proto Y Architecture
+========================================
+
+Overview
+****************************************
+
+The Proto Y Architecture is a framework for abstract prototype learning methods.
+
+It divides the problem into multiple steps:
+
+    * **Components** : Recalling the position and metadata of the components/prototypes.
+    * **Backbone** : Apply a mapping function to data and prototypes.
+    * **Comparison** : Calculate a dissimilarity based on the latent positions.
+    * **Competition** : Calculate competition values based on the comparison and the metadata.
+    * **Loss** : Calculate the loss based on the competition values
+    * **Inference** : Predict the output based on the competition values.
+
+Depending on the phase (Training or Testing) Loss or Inference is used.
+
+Inheritance Structure
+****************************************
+
+The Proto Y Architecture has a single base class that defines all steps and hooks
+of the architecture.
+
+.. autoclass:: prototorch.y.architectures.base.BaseYArchitecture
+
+    **Steps**
+
+    Components
+
+    .. automethod:: init_components
+    .. automethod:: components
+
+    Backbone
+
+    .. automethod:: init_backbone
+    .. automethod:: backbone
+
+    Comparison
+
+    .. automethod:: init_comparison
+    .. automethod:: comparison
+
+    Competition
+
+    .. automethod:: init_competition
+    .. automethod:: competition
+
+    Loss
+
+    .. automethod:: init_loss
+    .. automethod:: loss
+
+    Inference
+
+    .. automethod:: init_inference
+    .. automethod:: inference
+
+    **Hooks**
+
+    Torchmetric
+
+    .. automethod:: register_torchmetric
+
+Hyperparameters
+****************************************
+Every model implemented with the Proto Y Architecture has a set of hyperparameters,
+which is stored in the ``HyperParameters`` attribute of the architecture.

examples/cbc_iris.py

@@ -1,69 +0,0 @@
-"""CBC example using the Iris dataset."""
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import CBC, VisCBC2D
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Reproducibility
-    seed_everything(seed=4)
-
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    train_ds = pt.datasets.Iris(dims=[0, 2])
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=32)
-
-    # Hyperparameters
-    hparams = dict(
-        distribution=[1, 0, 3],
-        margin=0.1,
-        proto_lr=0.01,
-        bb_lr=0.01,
-    )
-
-    # Initialize the model
-    model = CBC(
-        hparams,
-        components_initializer=pt.initializers.SSCI(train_ds, noise=0.1),
-        reasonings_initializer=pt.initializers.
-        PurePositiveReasoningsInitializer(),
-    )
-
-    # Callbacks
-    vis = VisCBC2D(
-        data=train_ds,
-        title="CBC Iris Example",
-        resolution=100,
-        axis_off=True,
-    )
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-        ],
-        detect_anomaly=True,
-        log_every_n_steps=1,
-        max_epochs=1000,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/dynamic_pruning.py

@@ -1,102 +0,0 @@
-"""Dynamically prune 'loser' prototypes in GLVQ-type models."""
-
-import argparse
-import logging
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import (
-    CELVQ,
-    PruneLoserPrototypes,
-    VisGLVQ2D,
-)
-from pytorch_lightning.callbacks import EarlyStopping
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Reproducibility
-    seed_everything(seed=4)
-
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    num_classes = 4
-    num_features = 2
-    num_clusters = 1
-    train_ds = pt.datasets.Random(
-        num_samples=500,
-        num_classes=num_classes,
-        num_features=num_features,
-        num_clusters=num_clusters,
-        separation=3.0,
-        seed=42,
-    )
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=256)
-
-    # Hyperparameters
-    prototypes_per_class = num_clusters * 5
-    hparams = dict(
-        distribution=(num_classes, prototypes_per_class),
-        lr=0.2,
-    )
-
-    # Initialize the model
-    model = CELVQ(
-        hparams,
-        prototypes_initializer=pt.initializers.FVCI(2, 3.0),
-    )
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Summary
-    logging.info(model)
-
-    # Callbacks
-    vis = VisGLVQ2D(train_ds)
-    pruning = PruneLoserPrototypes(
-        threshold=0.01,  # prune prototype if it wins less than 1%
-        idle_epochs=20,  # pruning too early may cause problems
-        prune_quota_per_epoch=2,  # prune at most 2 prototypes per epoch
-        frequency=1,  # prune every epoch
-        verbose=True,
-    )
-    es = EarlyStopping(
-        monitor="train_loss",
-        min_delta=0.001,
-        patience=20,
-        mode="min",
-        verbose=True,
-        check_on_train_epoch_end=True,
-    )
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-            pruning,
-            es,
-        ],
-        detect_anomaly=True,
-        log_every_n_steps=1,
-        max_epochs=1000,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/glvq_iris.py

@@ -1,82 +0,0 @@
-"""GLVQ example using the Iris dataset."""
-
-import argparse
-import logging
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import GLVQ, VisGLVQ2D
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.optim.lr_scheduler import ExponentialLR
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=UserWarning)
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-
-if __name__ == "__main__":
-    # Reproducibility
-    seed_everything(seed=4)
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    train_ds = pt.datasets.Iris(dims=[0, 2])
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=64, num_workers=4)
-
-    # Hyperparameters
-    hparams = dict(
-        distribution={
-            "num_classes": 3,
-            "per_class": 4
-        },
-        lr=0.01,
-    )
-
-    # Initialize the model
-    model = GLVQ(
-        hparams,
-        optimizer=torch.optim.Adam,
-        prototypes_initializer=pt.initializers.SMCI(train_ds),
-        lr_scheduler=ExponentialLR,
-        lr_scheduler_kwargs=dict(gamma=0.99, verbose=False),
-    )
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Callbacks
-    vis = VisGLVQ2D(data=train_ds)
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-        ],
-        max_epochs=100,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)
-
-    # Manual save
-    trainer.save_checkpoint("./glvq_iris.ckpt")
-
-    # Load saved model
-    new_model = GLVQ.load_from_checkpoint(
-        checkpoint_path="./glvq_iris.ckpt",
-        strict=False,
-    )
-    logging.info(new_model)

examples/gmlvq_iris.py

@@ -1,78 +1,144 @@
-"""GMLVQ example using the Iris dataset."""
+import logging
 
-import argparse
-import warnings
-
-import prototorch as pt
 import pytorch_lightning as pl
-import torch
+import torchmetrics
-from lightning_fabric.utilities.seed import seed_everything
+from prototorch.core import SMCI, PCALinearTransformInitializer
-from prototorch.models import GMLVQ, VisGMLVQ2D
+from prototorch.datasets import Iris
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
+from prototorch.models.architectures.base import Steps
-from torch.optim.lr_scheduler import ExponentialLR
+from prototorch.models.callbacks import (
-from torch.utils.data import DataLoader
+    LogTorchmetricCallback,
+    PlotLambdaMatrixToTensorboard,
+    VisGMLVQ2D,
+)
+from prototorch.models.library.gmlvq import GMLVQ
+from pytorch_lightning.callbacks import EarlyStopping
+from torch.utils.data import DataLoader, random_split
 
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
+logging.basicConfig(level=logging.INFO)
-warnings.filterwarnings("ignore", category=UserWarning)
 
-if __name__ == "__main__":
+# ##############################################################################
 
-    # Reproducibility
-    seed_everything(seed=4)
 
-    # Command-line arguments
+def main():
-    parser = argparse.ArgumentParser()
+    # ------------------------------------------------------------
-    parser.add_argument("--gpus", type=int, default=0)
+    # DATA
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
+    # ------------------------------------------------------------
-    args = parser.parse_args()
 
     # Dataset
-    train_ds = pt.datasets.Iris()
+    full_dataset = Iris()
+    full_count = len(full_dataset)
 
-    # Dataloaders
+    train_count = int(full_count * 0.5)
-    train_loader = DataLoader(train_ds, batch_size=64)
+    val_count = int(full_count * 0.4)
+    test_count = int(full_count * 0.1)
 
-    # Hyperparameters
+    train_dataset, val_dataset, test_dataset = random_split(
-    hparams = dict(
+        full_dataset, (train_count, val_count, test_count))
+
+    # Dataloader
+    train_loader = DataLoader(
+        train_dataset,
+        batch_size=1,
+        num_workers=4,
+        shuffle=True,
+    )
+    val_loader = DataLoader(
+        val_dataset,
+        batch_size=1,
+        num_workers=4,
+        shuffle=False,
+    )
+    test_loader = DataLoader(
+        test_dataset,
+        batch_size=1,
+        num_workers=0,
+        shuffle=False,
+    )
+
+    # ------------------------------------------------------------
+    # HYPERPARAMETERS
+    # ------------------------------------------------------------
+
+    # Select Initializer
+    components_initializer = SMCI(full_dataset)
+
+    # Define Hyperparameters
+    hyperparameters = GMLVQ.HyperParameters(
+        lr=dict(components_layer=0.1, _omega=0),
         input_dim=4,
-        latent_dim=4,
+        distribution=dict(
-        distribution={
+            num_classes=3,
-            "num_classes": 3,
+            per_class=1,
-            "per_class": 2
+        ),
-        },
+        component_initializer=components_initializer,
-        proto_lr=0.01,
+        omega_initializer=PCALinearTransformInitializer,
-        bb_lr=0.01,
+        omega_initializer_kwargs=dict(
+            data=train_dataset.dataset[train_dataset.indices][0]))
+
+    # Create Model
+    model = GMLVQ(hyperparameters)
+
+    # ------------------------------------------------------------
+    # TRAINING
+    # ------------------------------------------------------------
+
+    # Controlling Callbacks
+    recall = LogTorchmetricCallback(
+        'training_recall',
+        torchmetrics.Recall,
+        num_classes=3,
+        step=Steps.TRAINING,
     )
 
-    # Initialize the model
+    stopping_criterion = LogTorchmetricCallback(
-    model = GMLVQ(
+        'validation_recall',
-        hparams,
+        torchmetrics.Recall,
-        optimizer=torch.optim.Adam,
+        num_classes=3,
-        prototypes_initializer=pt.initializers.SMCI(train_ds),
+        step=Steps.VALIDATION,
-        lr_scheduler=ExponentialLR,
-        lr_scheduler_kwargs=dict(gamma=0.99, verbose=False),
     )
 
-    # Compute intermediate input and output sizes
+    accuracy = LogTorchmetricCallback(
-    model.example_input_array = torch.zeros(4, 4)
+        'validation_accuracy',
+        torchmetrics.Accuracy,
+        num_classes=3,
+        step=Steps.VALIDATION,
+    )
 
-    # Callbacks
+    es = EarlyStopping(
-    vis = VisGMLVQ2D(data=train_ds)
+        monitor=stopping_criterion.name,
+        mode="max",
+        patience=10,
+    )
 
-    # Setup trainer
+    # Visualization Callback
+    vis = VisGMLVQ2D(data=full_dataset)
+
+    # Define trainer
     trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
         callbacks=[
             vis,
+            recall,
+            accuracy,
+            stopping_criterion,
+            es,
+            PlotLambdaMatrixToTensorboard(),
         ],
         max_epochs=100,
-        log_every_n_steps=1,
-        detect_anomaly=True,
     )
 
-    # Training loop
+    # Train
-    trainer.fit(model, train_loader)
+    trainer.fit(model, train_loader, val_loader)
+    trainer.test(model, test_loader)
 
-    torch.save(model, "iris.pth")
+    # Manual save
+    trainer.save_checkpoint("./y_arch.ckpt")
+
+    # Load saved model
+    new_model = GMLVQ.load_from_checkpoint(
+        checkpoint_path="./y_arch.ckpt",
+        strict=True,
+    )
+
+
+if __name__ == "__main__":
+    main()

examples/gmlvq_mnist.py

@@ -1,115 +0,0 @@
-"""GMLVQ example using the MNIST dataset."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import (
-    ImageGMLVQ,
-    PruneLoserPrototypes,
-    VisImgComp,
-)
-from pytorch_lightning.callbacks import EarlyStopping
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-from torchvision import transforms
-from torchvision.datasets import MNIST
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Reproducibility
-    seed_everything(seed=4)
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    train_ds = MNIST(
-        "~/datasets",
-        train=True,
-        download=True,
-        transform=transforms.Compose([
-            transforms.ToTensor(),
-        ]),
-    )
-    test_ds = MNIST(
-        "~/datasets",
-        train=False,
-        download=True,
-        transform=transforms.Compose([
-            transforms.ToTensor(),
-        ]),
-    )
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, num_workers=4, batch_size=256)
-    test_loader = DataLoader(test_ds, num_workers=4, batch_size=256)
-
-    # Hyperparameters
-    num_classes = 10
-    prototypes_per_class = 10
-    hparams = dict(
-        input_dim=28 * 28,
-        latent_dim=28 * 28,
-        distribution=(num_classes, prototypes_per_class),
-        proto_lr=0.01,
-        bb_lr=0.01,
-    )
-
-    # Initialize the model
-    model = ImageGMLVQ(
-        hparams,
-        optimizer=torch.optim.Adam,
-        prototypes_initializer=pt.initializers.SMCI(train_ds),
-    )
-
-    # Callbacks
-    vis = VisImgComp(
-        data=train_ds,
-        num_columns=10,
-        show=False,
-        tensorboard=True,
-        random_data=100,
-        add_embedding=True,
-        embedding_data=200,
-        flatten_data=False,
-    )
-    pruning = PruneLoserPrototypes(
-        threshold=0.01,
-        idle_epochs=1,
-        prune_quota_per_epoch=10,
-        frequency=1,
-        verbose=True,
-    )
-    es = EarlyStopping(
-        monitor="train_loss",
-        min_delta=0.001,
-        patience=15,
-        mode="min",
-        check_on_train_epoch_end=True,
-    )
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-            pruning,
-            es,
-        ],
-        max_epochs=1000,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/gmlvq_spiral.py

@@ -1,97 +0,0 @@
-"""GMLVQ example using the spiral dataset."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import (
-    GMLVQ,
-    PruneLoserPrototypes,
-    VisGLVQ2D,
-)
-from pytorch_lightning.callbacks import EarlyStopping
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Reproducibility
-    seed_everything(seed=4)
-
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    train_ds = pt.datasets.Spiral(num_samples=500, noise=0.5)
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=256)
-
-    # Hyperparameters
-    num_classes = 2
-    prototypes_per_class = 10
-    hparams = dict(
-        distribution=(num_classes, prototypes_per_class),
-        transfer_function="swish_beta",
-        transfer_beta=10.0,
-        proto_lr=0.1,
-        bb_lr=0.1,
-        input_dim=2,
-        latent_dim=2,
-    )
-
-    # Initialize the model
-    model = GMLVQ(
-        hparams,
-        optimizer=torch.optim.Adam,
-        prototypes_initializer=pt.initializers.SSCI(train_ds, noise=1e-2),
-    )
-
-    # Callbacks
-    vis = VisGLVQ2D(
-        train_ds,
-        show_last_only=False,
-        block=False,
-    )
-    pruning = PruneLoserPrototypes(
-        threshold=0.01,
-        idle_epochs=10,
-        prune_quota_per_epoch=5,
-        frequency=5,
-        replace=True,
-        prototypes_initializer=pt.initializers.SSCI(train_ds, noise=1e-1),
-        verbose=True,
-    )
-    es = EarlyStopping(
-        monitor="train_loss",
-        min_delta=1.0,
-        patience=5,
-        mode="min",
-        check_on_train_epoch_end=True,
-    )
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-            es,
-            pruning,
-        ],
-        max_epochs=1000,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/gng_iris.py

@@ -1,68 +0,0 @@
-"""Growing Neural Gas example using the Iris dataset."""
-
-import argparse
-import logging
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import GrowingNeuralGas, VisNG2D
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Reproducibility
-    seed_everything(seed=42)
-
-    # Prepare the data
-    train_ds = pt.datasets.Iris(dims=[0, 2])
-    train_loader = DataLoader(train_ds, batch_size=64)
-
-    # Hyperparameters
-    hparams = dict(
-        num_prototypes=5,
-        input_dim=2,
-        lr=0.1,
-    )
-
-    # Initialize the model
-    model = GrowingNeuralGas(
-        hparams,
-        prototypes_initializer=pt.initializers.ZCI(2),
-    )
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Model summary
-    logging.info(model)
-
-    # Callbacks
-    vis = VisNG2D(data=train_loader)
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-        ],
-        max_epochs=100,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/grlvq_iris.py

@@ -1,77 +0,0 @@
-"""GMLVQ example using the Iris dataset."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import GRLVQ, VisSiameseGLVQ2D
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.optim.lr_scheduler import ExponentialLR
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-
-    # Reproducibility
-    seed_everything(seed=4)
-
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    train_ds = pt.datasets.Iris([0, 1])
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=64)
-
-    # Hyperparameters
-    hparams = dict(
-        input_dim=2,
-        distribution={
-            "num_classes": 3,
-            "per_class": 2
-        },
-        proto_lr=0.01,
-        bb_lr=0.01,
-    )
-
-    # Initialize the model
-    model = GRLVQ(
-        hparams,
-        optimizer=torch.optim.Adam,
-        prototypes_initializer=pt.initializers.SMCI(train_ds),
-        lr_scheduler=ExponentialLR,
-        lr_scheduler_kwargs=dict(gamma=0.99, verbose=False),
-    )
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Callbacks
-    vis = VisSiameseGLVQ2D(data=train_ds)
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-        ],
-        max_epochs=5,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)
-
-    torch.save(model, "iris.pth")

examples/gtlvq_mnist.py

@@ -1,119 +0,0 @@
-"""GTLVQ example using the MNIST dataset."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import (
-    ImageGTLVQ,
-    PruneLoserPrototypes,
-    VisImgComp,
-)
-from pytorch_lightning.callbacks import EarlyStopping
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-from torchvision import transforms
-from torchvision.datasets import MNIST
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Reproducibility
-    seed_everything(seed=4)
-
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    train_ds = MNIST(
-        "~/datasets",
-        train=True,
-        download=True,
-        transform=transforms.Compose([
-            transforms.ToTensor(),
-        ]),
-    )
-    test_ds = MNIST(
-        "~/datasets",
-        train=False,
-        download=True,
-        transform=transforms.Compose([
-            transforms.ToTensor(),
-        ]),
-    )
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, num_workers=0, batch_size=256)
-    test_loader = DataLoader(test_ds, num_workers=0, batch_size=256)
-
-    # Hyperparameters
-    num_classes = 10
-    prototypes_per_class = 1
-    hparams = dict(
-        input_dim=28 * 28,
-        latent_dim=28,
-        distribution=(num_classes, prototypes_per_class),
-        proto_lr=0.01,
-        bb_lr=0.01,
-    )
-
-    # Initialize the model
-    model = ImageGTLVQ(
-        hparams,
-        optimizer=torch.optim.Adam,
-        prototypes_initializer=pt.initializers.SMCI(train_ds),
-        #Use one batch of data for subspace initiator.
-        omega_initializer=pt.initializers.PCALinearTransformInitializer(
-            next(iter(train_loader))[0].reshape(256, 28 * 28)))
-
-    # Callbacks
-    vis = VisImgComp(
-        data=train_ds,
-        num_columns=10,
-        show=False,
-        tensorboard=True,
-        random_data=100,
-        add_embedding=True,
-        embedding_data=200,
-        flatten_data=False,
-    )
-    pruning = PruneLoserPrototypes(
-        threshold=0.01,
-        idle_epochs=1,
-        prune_quota_per_epoch=10,
-        frequency=1,
-        verbose=True,
-    )
-    es = EarlyStopping(
-        monitor="train_loss",
-        min_delta=0.001,
-        patience=15,
-        mode="min",
-        check_on_train_epoch_end=True,
-    )
-
-    # Setup trainer
-    # using GPUs here is strongly recommended!
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-            pruning,
-            es,
-        ],
-        max_epochs=1000,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/gtlvq_moons.py

@@ -1,79 +0,0 @@
-"""Localized-GTLVQ example using the Moons dataset."""
-
-import argparse
-import logging
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import GTLVQ, VisGLVQ2D
-from pytorch_lightning.callbacks import EarlyStopping
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Reproducibility
-    seed_everything(seed=2)
-
-    # Dataset
-    train_ds = pt.datasets.Moons(num_samples=300, noise=0.2, seed=42)
-
-    # Dataloaders
-    train_loader = DataLoader(
-        train_ds,
-        batch_size=256,
-        shuffle=True,
-    )
-
-    # Hyperparameters
-    # Latent_dim should be lower than input dim.
-    hparams = dict(distribution=[1, 3], input_dim=2, latent_dim=1)
-
-    # Initialize the model
-    model = GTLVQ(hparams,
-                  prototypes_initializer=pt.initializers.SMCI(train_ds))
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Summary
-    logging.info(model)
-
-    # Callbacks
-    vis = VisGLVQ2D(data=train_ds)
-    es = EarlyStopping(
-        monitor="train_acc",
-        min_delta=0.001,
-        patience=20,
-        mode="max",
-        verbose=False,
-        check_on_train_epoch_end=True,
-    )
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-            es,
-        ],
-        max_epochs=1000,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/knn_iris.py

@@ -1,84 +0,0 @@
-"""k-NN example using the Iris dataset from scikit-learn."""
-
-import argparse
-import logging
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from prototorch.models import KNN, VisGLVQ2D
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from sklearn.datasets import load_iris
-from sklearn.model_selection import train_test_split
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-
-if __name__ == "__main__":
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    X, y = load_iris(return_X_y=True)
-    X = X[:, 0:3:2]
-
-    X_train, X_test, y_train, y_test = train_test_split(
-        X,
-        y,
-        test_size=0.5,
-        random_state=42,
-    )
-
-    train_ds = pt.datasets.NumpyDataset(X_train, y_train)
-    test_ds = pt.datasets.NumpyDataset(X_test, y_test)
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=16)
-    test_loader = DataLoader(test_ds, batch_size=16)
-
-    # Hyperparameters
-    hparams = dict(k=5)
-
-    # Initialize the model
-    model = KNN(hparams, data=train_ds)
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Summary
-    logging.info(model)
-
-    # Callbacks
-    vis = VisGLVQ2D(
-        data=(X_train, y_train),
-        resolution=200,
-        block=True,
-    )
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        max_epochs=1,
-        callbacks=[
-            vis,
-        ],
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    # This is only for visualization. k-NN has no training phase.
-    trainer.fit(model, train_loader)
-
-    # Recall
-    y_pred = model.predict(torch.tensor(X_train))
-    logging.info(y_pred)
-
-    # Test
-    trainer.test(model, dataloaders=test_loader)

examples/ksom_colors.py

@@ -1,121 +0,0 @@
-"""Kohonen Self Organizing Map."""
-
-import argparse
-import logging
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from matplotlib import pyplot as plt
-from prototorch.models import KohonenSOM
-from prototorch.utils.colors import hex_to_rgb
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader, TensorDataset
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-
-class Vis2DColorSOM(pl.Callback):
-
-    def __init__(self, data, title="ColorSOMe", pause_time=0.1):
-        super().__init__()
-        self.title = title
-        self.fig = plt.figure(self.title)
-        self.data = data
-        self.pause_time = pause_time
-
-    def on_train_epoch_end(self, trainer, pl_module: KohonenSOM):
-        ax = self.fig.gca()
-        ax.cla()
-        ax.set_title(self.title)
-        h, w = pl_module._grid.shape[:2]
-        protos = pl_module.prototypes.view(h, w, 3)
-        ax.imshow(protos)
-        ax.axis("off")
-
-        # Overlay color names
-        d = pl_module.compute_distances(self.data)
-        wp = pl_module.predict_from_distances(d)
-        for i, iloc in enumerate(wp):
-            plt.text(
-                iloc[1],
-                iloc[0],
-                color_names[i],
-                ha="center",
-                va="center",
-                bbox=dict(facecolor="white", alpha=0.5, lw=0),
-            )
-
-        if trainer.current_epoch != trainer.max_epochs - 1:
-            plt.pause(self.pause_time)
-        else:
-            plt.show(block=True)
-
-
-if __name__ == "__main__":
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Reproducibility
-    seed_everything(seed=42)
-
-    # Prepare the data
-    hex_colors = [
-        "#000000", "#0000ff", "#00007f", "#1f86ff", "#5466aa", "#997fff",
-        "#00ff00", "#ff0000", "#00ffff", "#ff00ff", "#ffff00", "#ffffff",
-        "#545454", "#7f7f7f", "#a8a8a8", "#808000", "#800080", "#ffa500"
-    ]
-    color_names = [
-        "black", "blue", "darkblue", "skyblue", "greyblue", "lilac", "green",
-        "red", "cyan", "magenta", "yellow", "white", "darkgrey", "mediumgrey",
-        "lightgrey", "olive", "purple", "orange"
-    ]
-    colors = list(hex_to_rgb(hex_colors))
-    data = torch.Tensor(colors) / 255.0
-    train_ds = TensorDataset(data)
-    train_loader = DataLoader(train_ds, batch_size=8)
-
-    # Hyperparameters
-    hparams = dict(
-        shape=(18, 32),
-        alpha=1.0,
-        sigma=16,
-        lr=0.1,
-    )
-
-    # Initialize the model
-    model = KohonenSOM(
-        hparams,
-        prototypes_initializer=pt.initializers.RNCI(3),
-    )
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 3)
-
-    # Model summary
-    logging.info(model)
-
-    # Callbacks
-    vis = Vis2DColorSOM(data=data)
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        max_epochs=500,
-        callbacks=[
-            vis,
-        ],
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/lgmlvq_moons.py

@@ -1,80 +0,0 @@
-"""Localized-GMLVQ example using the Moons dataset."""
-
-import argparse
-import logging
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import LGMLVQ, VisGLVQ2D
-from pytorch_lightning.callbacks import EarlyStopping
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Reproducibility
-    seed_everything(seed=2)
-
-    # Dataset
-    train_ds = pt.datasets.Moons(num_samples=300, noise=0.2, seed=42)
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=256, shuffle=True)
-
-    # Hyperparameters
-    hparams = dict(
-        distribution=[1, 3],
-        input_dim=2,
-        latent_dim=2,
-    )
-
-    # Initialize the model
-    model = LGMLVQ(
-        hparams,
-        prototypes_initializer=pt.initializers.SMCI(train_ds),
-    )
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Summary
-    logging.info(model)
-
-    # Callbacks
-    vis = VisGLVQ2D(data=train_ds)
-    es = EarlyStopping(
-        monitor="train_acc",
-        min_delta=0.001,
-        patience=20,
-        mode="max",
-        verbose=False,
-        check_on_train_epoch_end=True,
-    )
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-            es,
-        ],
-        log_every_n_steps=1,
-        max_epochs=1000,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/lvqmln_iris.py

@@ -1,106 +0,0 @@
-"""LVQMLN example using all four dimensions of the Iris dataset."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import (
-    LVQMLN,
-    PruneLoserPrototypes,
-    VisSiameseGLVQ2D,
-)
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-
-class Backbone(torch.nn.Module):
-
-    def __init__(self, input_size=4, hidden_size=10, latent_size=2):
-        super().__init__()
-        self.input_size = input_size
-        self.hidden_size = hidden_size
-        self.latent_size = latent_size
-        self.dense1 = torch.nn.Linear(self.input_size, self.hidden_size)
-        self.dense2 = torch.nn.Linear(self.hidden_size, self.latent_size)
-        self.activation = torch.nn.Sigmoid()
-
-    def forward(self, x):
-        x = self.activation(self.dense1(x))
-        out = self.activation(self.dense2(x))
-        return out
-
-
-if __name__ == "__main__":
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    train_ds = pt.datasets.Iris()
-
-    # Reproducibility
-    seed_everything(seed=42)
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=150)
-
-    # Hyperparameters
-    hparams = dict(
-        distribution=[3, 4, 5],
-        proto_lr=0.001,
-        bb_lr=0.001,
-    )
-
-    # Initialize the backbone
-    backbone = Backbone()
-
-    # Initialize the model
-    model = LVQMLN(
-        hparams,
-        prototypes_initializer=pt.initializers.SSCI(
-            train_ds,
-            transform=backbone,
-        ),
-        backbone=backbone,
-    )
-
-    # Callbacks
-    vis = VisSiameseGLVQ2D(
-        data=train_ds,
-        map_protos=False,
-        border=0.1,
-        resolution=500,
-        axis_off=True,
-    )
-    pruning = PruneLoserPrototypes(
-        threshold=0.01,
-        idle_epochs=20,
-        prune_quota_per_epoch=2,
-        frequency=10,
-        verbose=True,
-    )
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-            pruning,
-        ],
-        log_every_n_steps=1,
-        max_epochs=1000,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/median_lvq_iris.py

@@ -1,71 +0,0 @@
-"""Median-LVQ example using the Iris dataset."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import MedianLVQ, VisGLVQ2D
-from pytorch_lightning.callbacks import EarlyStopping
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Reproducibility
-    seed_everything(seed=4)
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    train_ds = pt.datasets.Iris(dims=[0, 2])
-
-    # Dataloaders
-    train_loader = DataLoader(
-        train_ds,
-        batch_size=len(train_ds),  # MedianLVQ cannot handle mini-batches
-    )
-
-    # Initialize the model
-    model = MedianLVQ(
-        hparams=dict(distribution=(3, 2), lr=0.01),
-        prototypes_initializer=pt.initializers.SSCI(train_ds),
-    )
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Callbacks
-    vis = VisGLVQ2D(data=train_ds)
-    es = EarlyStopping(
-        monitor="train_acc",
-        min_delta=0.01,
-        patience=5,
-        mode="max",
-        verbose=True,
-        check_on_train_epoch_end=True,
-    )
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-            es,
-        ],
-        max_epochs=1000,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/ng_iris.py

@@ -1,77 +0,0 @@
-"""Neural Gas example using the Iris dataset."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import NeuralGas, VisNG2D
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from sklearn.datasets import load_iris
-from sklearn.preprocessing import StandardScaler
-from torch.optim.lr_scheduler import ExponentialLR
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Reproducibility
-    seed_everything(seed=4)
-
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Prepare and pre-process the dataset
-    x_train, y_train = load_iris(return_X_y=True)
-    x_train = x_train[:, 0:3:2]
-    scaler = StandardScaler()
-    scaler.fit(x_train)
-    x_train = scaler.transform(x_train)
-
-    train_ds = pt.datasets.NumpyDataset(x_train, y_train)
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=150)
-
-    # Hyperparameters
-    hparams = dict(
-        num_prototypes=30,
-        input_dim=2,
-        lr=0.03,
-    )
-
-    # Initialize the model
-    model = NeuralGas(
-        hparams,
-        prototypes_initializer=pt.core.ZCI(2),
-        lr_scheduler=ExponentialLR,
-        lr_scheduler_kwargs=dict(gamma=0.99, verbose=False),
-    )
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Callbacks
-    vis = VisNG2D(data=train_ds)
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-        ],
-        max_epochs=1000,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/rslvq_iris.py

@@ -1,71 +0,0 @@
-"""RSLVQ example using the Iris dataset."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import RSLVQ, VisGLVQ2D
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-if __name__ == "__main__":
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Reproducibility
-    seed_everything(seed=42)
-
-    # Dataset
-    train_ds = pt.datasets.Iris(dims=[0, 2])
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=64)
-
-    # Hyperparameters
-    hparams = dict(
-        distribution=[2, 2, 3],
-        proto_lr=0.05,
-        lambd=0.1,
-        variance=1.0,
-        input_dim=2,
-        latent_dim=2,
-        bb_lr=0.01,
-    )
-
-    # Initialize the model
-    model = RSLVQ(
-        hparams,
-        optimizer=torch.optim.Adam,
-        prototypes_initializer=pt.initializers.SSCI(train_ds, noise=0.2),
-    )
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Callbacks
-    vis = VisGLVQ2D(data=train_ds)
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-        ],
-        detect_anomaly=True,
-        max_epochs=100,
-        log_every_n_steps=1,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/siamese_glvq_iris.py

@@ -1,85 +0,0 @@
-"""Siamese GLVQ example using all four dimensions of the Iris dataset."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import SiameseGLVQ, VisSiameseGLVQ2D
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-
-class Backbone(torch.nn.Module):
-
-    def __init__(self, input_size=4, hidden_size=10, latent_size=2):
-        super().__init__()
-        self.input_size = input_size
-        self.hidden_size = hidden_size
-        self.latent_size = latent_size
-        self.dense1 = torch.nn.Linear(self.input_size, self.hidden_size)
-        self.dense2 = torch.nn.Linear(self.hidden_size, self.latent_size)
-        self.activation = torch.nn.Sigmoid()
-
-    def forward(self, x):
-        x = self.activation(self.dense1(x))
-        out = self.activation(self.dense2(x))
-        return out
-
-
-if __name__ == "__main__":
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    train_ds = pt.datasets.Iris()
-
-    # Reproducibility
-    seed_everything(seed=2)
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=150)
-
-    # Hyperparameters
-    hparams = dict(
-        distribution=[1, 2, 3],
-        lr=0.01,
-    )
-
-    # Initialize the backbone
-    backbone = Backbone()
-
-    # Initialize the model
-    model = SiameseGLVQ(
-        hparams,
-        prototypes_initializer=pt.initializers.SMCI(train_ds),
-        backbone=backbone,
-        both_path_gradients=False,
-    )
-
-    # Callbacks
-    vis = VisSiameseGLVQ2D(data=train_ds, border=0.1)
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-        ],
-        max_epochs=1000,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/siamese_gtlvq_iris.py

@@ -1,87 +0,0 @@
-"""Siamese GTLVQ example using all four dimensions of the Iris dataset."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import SiameseGTLVQ, VisSiameseGLVQ2D
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-warnings.filterwarnings("ignore", category=UserWarning)
-
-
-class Backbone(torch.nn.Module):
-
-    def __init__(self, input_size=4, hidden_size=10, latent_size=2):
-        super().__init__()
-        self.input_size = input_size
-        self.hidden_size = hidden_size
-        self.latent_size = latent_size
-        self.dense1 = torch.nn.Linear(self.input_size, self.hidden_size)
-        self.dense2 = torch.nn.Linear(self.hidden_size, self.latent_size)
-        self.activation = torch.nn.Sigmoid()
-
-    def forward(self, x):
-        x = self.activation(self.dense1(x))
-        out = self.activation(self.dense2(x))
-        return out
-
-
-if __name__ == "__main__":
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Dataset
-    train_ds = pt.datasets.Iris()
-
-    # Reproducibility
-    seed_everything(seed=2)
-
-    # Dataloaders
-    train_loader = DataLoader(train_ds, batch_size=150)
-
-    # Hyperparameters
-    hparams = dict(
-        distribution=[1, 2, 3],
-        lr=0.01,
-        input_dim=2,
-        latent_dim=1,
-    )
-
-    # Initialize the backbone
-    backbone = Backbone(latent_size=hparams["input_dim"])
-
-    # Initialize the model
-    model = SiameseGTLVQ(
-        hparams,
-        prototypes_initializer=pt.initializers.SMCI(train_ds),
-        backbone=backbone,
-        both_path_gradients=False,
-    )
-
-    # Callbacks
-    vis = VisSiameseGLVQ2D(data=train_ds, border=0.1)
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-        ],
-        max_epochs=1000,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

examples/warm_starting.py

@@ -1,129 +0,0 @@
-"""Warm-starting GLVQ with prototypes from Growing Neural Gas."""
-
-import argparse
-import warnings
-
-import prototorch as pt
-import pytorch_lightning as pl
-import torch
-from lightning_fabric.utilities.seed import seed_everything
-from prototorch.models import (
-    GLVQ,
-    KNN,
-    GrowingNeuralGas,
-    PruneLoserPrototypes,
-    VisGLVQ2D,
-)
-from pytorch_lightning.callbacks import EarlyStopping
-from pytorch_lightning.utilities.warnings import PossibleUserWarning
-from torch.optim.lr_scheduler import ExponentialLR
-from torch.utils.data import DataLoader
-
-warnings.filterwarnings("ignore", category=PossibleUserWarning)
-
-if __name__ == "__main__":
-
-    # Reproducibility
-    seed_everything(seed=4)
-    # Command-line arguments
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--gpus", type=int, default=0)
-    parser.add_argument("--fast_dev_run", type=bool, default=False)
-    args = parser.parse_args()
-
-    # Prepare the data
-    train_ds = pt.datasets.Iris(dims=[0, 2])
-    train_loader = DataLoader(train_ds, batch_size=64, num_workers=0)
-
-    # Initialize the gng
-    gng = GrowingNeuralGas(
-        hparams=dict(num_prototypes=5, insert_freq=2, lr=0.1),
-        prototypes_initializer=pt.initializers.ZCI(2),
-        lr_scheduler=ExponentialLR,
-        lr_scheduler_kwargs=dict(gamma=0.99, verbose=False),
-    )
-
-    # Callbacks
-    es = EarlyStopping(
-        monitor="loss",
-        min_delta=0.001,
-        patience=20,
-        mode="min",
-        verbose=False,
-        check_on_train_epoch_end=True,
-    )
-
-    # Setup trainer for GNG
-    trainer = pl.Trainer(
-        accelerator="cpu",
-        max_epochs=50 if args.fast_dev_run else
-        1000,  # 10 epochs fast dev run reproducible DIV error.
-        callbacks=[
-            es,
-        ],
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(gng, train_loader)
-
-    # Hyperparameters
-    hparams = dict(
-        distribution=[],
-        lr=0.01,
-    )
-
-    # Warm-start prototypes
-    knn = KNN(dict(k=1), data=train_ds)
-    prototypes = gng.prototypes
-    plabels = knn.predict(prototypes)
-
-    # Initialize the model
-    model = GLVQ(
-        hparams,
-        optimizer=torch.optim.Adam,
-        prototypes_initializer=pt.initializers.LCI(prototypes),
-        labels_initializer=pt.initializers.LLI(plabels),
-        lr_scheduler=ExponentialLR,
-        lr_scheduler_kwargs=dict(gamma=0.99, verbose=False),
-    )
-
-    # Compute intermediate input and output sizes
-    model.example_input_array = torch.zeros(4, 2)
-
-    # Callbacks
-    vis = VisGLVQ2D(data=train_ds)
-    pruning = PruneLoserPrototypes(
-        threshold=0.02,
-        idle_epochs=2,
-        prune_quota_per_epoch=5,
-        frequency=1,
-        verbose=True,
-    )
-    es = EarlyStopping(
-        monitor="train_loss",
-        min_delta=0.001,
-        patience=10,
-        mode="min",
-        verbose=True,
-        check_on_train_epoch_end=True,
-    )
-
-    # Setup trainer
-    trainer = pl.Trainer(
-        accelerator="cuda" if args.gpus else "cpu",
-        devices=args.gpus if args.gpus else "auto",
-        fast_dev_run=args.fast_dev_run,
-        callbacks=[
-            vis,
-            pruning,
-            es,
-        ],
-        max_epochs=1000,
-        log_every_n_steps=1,
-        detect_anomaly=True,
-    )
-
-    # Training loop
-    trainer.fit(model, train_loader)

prototorch/models/__init__.py

@@ -0,0 +1,25 @@
+from .architectures.base import BaseYArchitecture
+from .architectures.comparison import (
+    OmegaComparisonMixin,
+    SimpleComparisonMixin,
+)
+from .architectures.competition import WTACompetitionMixin
+from .architectures.components import SupervisedArchitecture
+from .architectures.loss import GLVQLossMixin
+from .architectures.optimization import (
+    MultipleLearningRateMixin,
+    SingleLearningRateMixin,
+)
+
+__all__ = [
+    'BaseYArchitecture',
+    "OmegaComparisonMixin",
+    "SimpleComparisonMixin",
+    "SingleLearningRateMixin",
+    "MultipleLearningRateMixin",
+    "SupervisedArchitecture",
+    "WTACompetitionMixin",
+    "GLVQLossMixin",
+]
+
+__version__ = "1.0.0-a8"

prototorch/models/architectures/base.py

@@ -0,0 +1,290 @@
+"""
+Proto Y Architecture
+
+Network architecture for Component based Learning.
+"""
+from __future__ import annotations
+
+from dataclasses import asdict, dataclass
+from typing import Any, Callable
+
+import pytorch_lightning as pl
+import torch
+from torchmetrics import Metric
+
+
+class Steps(enumerate):
+    TRAINING = "training"
+    VALIDATION = "validation"
+    TEST = "test"
+    PREDICT = "predict"
+
+
+class BaseYArchitecture(pl.LightningModule):
+
+    @dataclass
+    class HyperParameters:
+        """
+        Add all hyperparameters in the inherited class.
+        """
+        ...
+
+    # Fields
+    registered_metrics: dict[str, dict[type[Metric], Metric]] = {
+        Steps.TRAINING: {},
+        Steps.VALIDATION: {},
+        Steps.TEST: {},
+    }
+    registered_metric_callbacks: dict[str, dict[type[Metric],
+                                                set[Callable]]] = {
+                                                    Steps.TRAINING: {},
+                                                    Steps.VALIDATION: {},
+                                                    Steps.TEST: {},
+                                                }
+
+    # Type Hints for Necessary Fields
+    components_layer: torch.nn.Module
+
+    def __init__(self, hparams) -> None:
+        if isinstance(hparams, dict):
+            self.save_hyperparameters(hparams)
+            # TODO: => Move into Component Child
+            del hparams["initialized_proto_shape"]
+            hparams = self.HyperParameters(**hparams)
+        else:
+            hparams_dict = asdict(hparams)
+            hparams_dict["component_initializer"] = None
+            self.save_hyperparameters(hparams_dict, )
+
+        super().__init__()
+
+        # Common Steps
+        self.init_components(hparams)
+        self.init_backbone(hparams)
+        self.init_comparison(hparams)
+        self.init_competition(hparams)
+
+        # Train Steps
+        self.init_loss(hparams)
+
+        # Inference Steps
+        self.init_inference(hparams)
+
+    # external API
+    def get_competition(self, batch, components):
+        '''
+        Returns the output of the competition layer.
+        '''
+        latent_batch, latent_components = self.backbone(batch, components)
+        # TODO: => Latent Hook
+        comparison_tensor = self.comparison(latent_batch, latent_components)
+        # TODO: => Comparison Hook
+        return comparison_tensor
+
+    def forward(self, batch):
+        '''
+        Returns the prediction.
+        '''
+        if isinstance(batch, torch.Tensor):
+            batch = (batch, None)
+        # TODO: manage different datatypes?
+        components = self.components_layer()
+        # TODO: => Component Hook
+        comparison_tensor = self.get_competition(batch, components)
+        # TODO: => Competition Hook
+        return self.inference(comparison_tensor, components)
+
+    def predict(self, batch):
+        """
+        Alias for forward
+        """
+        return self.forward(batch)
+
+    def forward_comparison(self, batch):
+        '''
+        Returns the Output of the comparison layer.
+        '''
+        if isinstance(batch, torch.Tensor):
+            batch = (batch, None)
+        # TODO: manage different datatypes?
+        components = self.components_layer()
+        # TODO: => Component Hook
+        return self.get_competition(batch, components)
+
+    def loss_forward(self, batch):
+        '''
+        Returns the output of the loss layer.
+        '''
+        # TODO: manage different datatypes?
+        components = self.components_layer()
+        # TODO: => Component Hook
+        comparison_tensor = self.get_competition(batch, components)
+        # TODO: => Competition Hook
+        return self.loss(comparison_tensor, batch, components)
+
+    # Empty Initialization
+    def init_components(self, hparams: HyperParameters) -> None:
+        """
+        All initialization necessary for the components step.
+        """
+
+    def init_backbone(self, hparams: HyperParameters) -> None:
+        """
+        All initialization necessary for the backbone step.
+        """
+
+    def init_comparison(self, hparams: HyperParameters) -> None:
+        """
+        All initialization necessary for the comparison step.
+        """
+
+    def init_competition(self, hparams: HyperParameters) -> None:
+        """
+        All initialization necessary for the competition step.
+        """
+
+    def init_loss(self, hparams: HyperParameters) -> None:
+        """
+        All initialization necessary for the loss step.
+        """
+
+    def init_inference(self, hparams: HyperParameters) -> None:
+        """
+        All initialization necessary for the inference step.
+        """
+
+    # Empty Steps
+    def components(self):
+        """
+        This step has no input.
+
+        It returns the components.
+        """
+        raise NotImplementedError(
+            "The components step has no reasonable default.")
+
+    def backbone(self, batch, components):
+        """
+        The backbone step receives the data batch and the components.
+        It can transform both by an arbitrary function.
+
+        It returns the transformed batch and components,
+        each of the same length as the original input.
+        """
+        return batch, components
+
+    def comparison(self, batch, components):
+        """
+        Takes a batch of size N and the component set of size M.
+
+        It returns an NxMxD tensor containing D (usually 1) pairwise comparison measures.
+        """
+        raise NotImplementedError(
+            "The comparison step has no reasonable default.")
+
+    def competition(self, comparison_measures, components):
+        """
+        Takes the tensor of comparison measures.
+
+        Assigns a competition vector to each class.
+        """
+        raise NotImplementedError(
+            "The competition step has no reasonable default.")
+
+    def loss(self, comparison_measures, batch, components):
+        """
+        Takes the tensor of competition measures.
+
+        Calculates a single loss value
+        """
+        raise NotImplementedError("The loss step has no reasonable default.")
+
+    def inference(self, comparison_measures, components):
+        """
+        Takes the tensor of competition measures.
+
+        Returns the inferred vector.
+        """
+        raise NotImplementedError(
+            "The inference step has no reasonable default.")
+
+    # Y Architecture Hooks
+
+    # internal API, called by models and callbacks
+    def register_torchmetric(
+        self,
+        name: Callable,
+        metric: type[Metric],
+        step: str = Steps.TRAINING,
+        **metric_kwargs,
+    ):
+        '''
+        Register a callback for evaluating a torchmetric.
+        '''
+        if step == Steps.PREDICT:
+            raise ValueError("Prediction metrics are not supported.")
+
+        if metric not in self.registered_metrics:
+            self.registered_metrics[step][metric] = metric(**metric_kwargs)
+            self.registered_metric_callbacks[step][metric] = {name}
+        else:
+            self.registered_metric_callbacks[step][metric].add(name)
+
+    def update_metrics_step(self, batch, step):
+        # Prediction Metrics
+        preds = self(batch)
+
+        _, y = batch
+        for metric in self.registered_metrics[step]:
+            instance = self.registered_metrics[step][metric].to(self.device)
+            instance(y, preds.reshape(y.shape))
+
+    def update_metrics_epoch(self, step):
+        for metric in self.registered_metrics[step]:
+            instance = self.registered_metrics[step][metric].to(self.device)
+            value = instance.compute()
+
+            for callback in self.registered_metric_callbacks[step][metric]:
+                callback(value, self)
+
+            instance.reset()
+
+    # Lightning steps
+    # -------------------------------------------------------------------------
+    # >>>> Training
+    def training_step(self, batch, batch_idx, optimizer_idx=None):
+        self.update_metrics_step(batch, Steps.TRAINING)
+
+        return self.loss_forward(batch)
+
+    def training_epoch_end(self, outputs) -> None:
+        self.update_metrics_epoch(Steps.TRAINING)
+
+    # >>>> Validation
+    def validation_step(self, batch, batch_idx):
+        self.update_metrics_step(batch, Steps.VALIDATION)
+
+        return self.loss_forward(batch)
+
+    def validation_epoch_end(self, outputs) -> None:
+        self.update_metrics_epoch(Steps.VALIDATION)
+
+    # >>>> Test
+    def test_step(self, batch, batch_idx):
+        self.update_metrics_step(batch, Steps.TEST)
+        return self.loss_forward(batch)
+
+    def test_epoch_end(self, outputs) -> None:
+        self.update_metrics_epoch(Steps.TEST)
+
+    # >>>> Prediction
+    def predict_step(self, batch, batch_idx, dataloader_idx=0):
+        return self.predict(batch)
+
+    # Check points
+    def on_save_checkpoint(self, checkpoint: dict[str, Any]) -> None:
+        # Compatible with Lightning
+        checkpoint["hyper_parameters"] = {
+            'hparams': checkpoint["hyper_parameters"]
+        }
+        return super().on_save_checkpoint(checkpoint)

prototorch/models/architectures/comparison.py

@@ -0,0 +1,148 @@
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+from typing import Callable
+
+import torch
+from prototorch.core.distances import euclidean_distance
+from prototorch.core.initializers import (
+    AbstractLinearTransformInitializer,
+    EyeLinearTransformInitializer,
+)
+from prototorch.models.architectures.base import BaseYArchitecture
+from prototorch.nn.wrappers import LambdaLayer
+from torch import Tensor
+from torch.nn.parameter import Parameter
+
+
+class SimpleComparisonMixin(BaseYArchitecture):
+    """
+    Simple Comparison
+
+    A comparison layer that only uses the positions of the components
+    and the batch for dissimilarity computation.
+    """
+
+    # HyperParameters
+    # ----------------------------------------------------------------------------------------------
+    @dataclass
+    class HyperParameters(BaseYArchitecture.HyperParameters):
+        """
+        comparison_fn: The comparison / dissimilarity function to use. Default: euclidean_distance.
+        comparison_args: Keyword arguments for the comparison function. Default: {}.
+        """
+        comparison_fn: Callable = euclidean_distance
+        comparison_args: dict = field(default_factory=dict)
+
+        comparison_parameters: dict = field(default_factory=dict)
+
+    # Steps
+    # ----------------------------------------------------------------------------------------------
+    def init_comparison(self, hparams: HyperParameters):
+        self.comparison_layer = LambdaLayer(
+            fn=hparams.comparison_fn,
+            **hparams.comparison_args,
+        )
+
+        self.comparison_kwargs: dict[str, Tensor] = {}
+
+    def comparison(self, batch, components):
+        comp_tensor, _ = components
+        batch_tensor, _ = batch
+
+        comp_tensor = comp_tensor.unsqueeze(1)
+
+        distances = self.comparison_layer(
+            batch_tensor,
+            comp_tensor,
+            **self.comparison_kwargs,
+        )
+
+        return distances
+
+
+class OmegaComparisonMixin(SimpleComparisonMixin):
+    """
+    Omega Comparison
+
+    A comparison layer that uses the positions of the components
+    and the batch for dissimilarity computation.
+    """
+
+    _omega: torch.Tensor
+
+    # HyperParameters
+    # ----------------------------------------------------------------------------------------------
+    @dataclass
+    class HyperParameters(SimpleComparisonMixin.HyperParameters):
+        """
+        input_dim: Necessary Field: The dimensionality of the input.
+        latent_dim:
+            The dimensionality of the latent space. Default: 2.
+        omega_initializer:
+            The initializer to use for the omega matrix. Default: EyeLinearTransformInitializer.
+        """
+        input_dim: int | None = None
+        latent_dim: int = 2
+        omega_initializer: type[
+            AbstractLinearTransformInitializer] = EyeLinearTransformInitializer
+        omega_initializer_kwargs: dict = field(default_factory=dict)
+
+    # Steps
+    # ----------------------------------------------------------------------------------------------
+    def init_comparison(self, hparams: HyperParameters) -> None:
+        super().init_comparison(hparams)
+
+        # Initialize the omega matrix
+        if hparams.input_dim is None:
+            raise ValueError("input_dim must be specified.")
+        else:
+            omega = hparams.omega_initializer(
+                **hparams.omega_initializer_kwargs).generate(
+                    hparams.input_dim,
+                    hparams.latent_dim,
+                )
+            self.register_parameter("_omega", Parameter(omega))
+            self.comparison_kwargs = dict(omega=self._omega)
+
+    # Properties
+    # ----------------------------------------------------------------------------------------------
+    @property
+    def omega_matrix(self):
+        '''
+        Omega Matrix. Mapping applied to data and prototypes.
+        '''
+        return self._omega.detach().cpu()
+
+    @property
+    def lambda_matrix(self):
+        '''
+        Lambda Matrix.
+        '''
+        omega = self._omega.detach()
+        lam = omega @ omega.T
+        return lam.detach().cpu()
+
+    @property
+    def relevance_profile(self):
+        '''
+        Relevance Profile. Main Diagonal of the Lambda Matrix.
+        '''
+        return self.lambda_matrix.diag().abs()
+
+    @property
+    def classification_influence_profile(self):
+        '''
+        Classification Influence Profile. Influence of each dimension.
+        '''
+        lam = self.lambda_matrix
+        return lam.abs().sum(0)
+
+    @property
+    def parameter_omega(self):
+        return self._omega
+
+    @parameter_omega.setter
+    def parameter_omega(self, new_omega):
+        with torch.no_grad():
+            self._omega.data.copy_(new_omega)

prototorch/models/architectures/competition.py

@@ -0,0 +1,29 @@
+from dataclasses import dataclass
+
+from prototorch.core.competitions import WTAC
+from prototorch.models.architectures.base import BaseYArchitecture
+
+
+class WTACompetitionMixin(BaseYArchitecture):
+    """
+    Winner Take All Competition
+
+    A competition layer that uses the winner-take-all strategy.
+    """
+
+    # HyperParameters
+    # ----------------------------------------------------------------------------------------------------
+    @dataclass
+    class HyperParameters(BaseYArchitecture.HyperParameters):
+        """
+        No hyperparameters.
+        """
+
+    # Steps
+    # ----------------------------------------------------------------------------------------------------
+    def init_inference(self, hparams: HyperParameters):
+        self.competition_layer = WTAC()
+
+    def inference(self, comparison_measures, components):
+        comp_labels = components[1]
+        return self.competition_layer(comparison_measures, comp_labels)

prototorch/models/architectures/components.py

@@ -0,0 +1,64 @@
+from dataclasses import dataclass
+
+from prototorch.core.components import LabeledComponents
+from prototorch.core.initializers import (
+    AbstractComponentsInitializer,
+    LabelsInitializer,
+    ZerosCompInitializer,
+)
+from prototorch.models import BaseYArchitecture
+
+
+class SupervisedArchitecture(BaseYArchitecture):
+    """
+    Supervised Architecture
+
+    An architecture that uses labeled Components as component Layer.
+    """
+    components_layer: LabeledComponents
+
+    # HyperParameters
+    # ----------------------------------------------------------------------------------------------------
+    @dataclass
+    class HyperParameters:
+        """
+        distribution: A valid prototype distribution. No default possible.
+        components_initializer: An implementation of AbstractComponentsInitializer. No default possible.
+        """
+        distribution: "dict[str, int]"
+        component_initializer: AbstractComponentsInitializer
+
+    # Steps
+    # ----------------------------------------------------------------------------------------------------
+    def init_components(self, hparams: HyperParameters):
+        if hparams.component_initializer is not None:
+            self.components_layer = LabeledComponents(
+                distribution=hparams.distribution,
+                components_initializer=hparams.component_initializer,
+                labels_initializer=LabelsInitializer(),
+            )
+            proto_shape = self.components_layer.components.shape[1:]
+            self.hparams["initialized_proto_shape"] = proto_shape
+        else:
+            # when restoring a checkpointed model
+            self.components_layer = LabeledComponents(
+                distribution=hparams.distribution,
+                components_initializer=ZerosCompInitializer(
+                    self.hparams["initialized_proto_shape"]),
+            )
+
+    # Properties
+    # ----------------------------------------------------------------------------------------------------
+    @property
+    def prototypes(self):
+        """
+        Returns the position of the prototypes.
+        """
+        return self.components_layer.components.detach().cpu()
+
+    @property
+    def prototype_labels(self):
+        """
+        Returns the labels of the prototypes.
+        """
+        return self.components_layer.labels.detach().cpu()

prototorch/models/architectures/loss.py

@@ -0,0 +1,42 @@
+from dataclasses import dataclass, field
+
+from prototorch.core.losses import GLVQLoss
+from prototorch.models.architectures.base import BaseYArchitecture
+
+
+class GLVQLossMixin(BaseYArchitecture):
+    """
+    GLVQ Loss
+
+    A loss layer that uses the Generalized Learning Vector Quantization (GLVQ) loss.
+    """
+
+    # HyperParameters
+    # ----------------------------------------------------------------------------------------------------
+    @dataclass
+    class HyperParameters(BaseYArchitecture.HyperParameters):
+        """
+        margin: The margin of the GLVQ loss. Default: 0.0.
+        transfer_fn: Transfer function to use. Default: sigmoid_beta.
+        transfer_args: Keyword arguments for the transfer function. Default: {beta: 10.0}.
+        """
+        margin: float = 0.0
+
+        transfer_fn: str = "sigmoid_beta"
+        transfer_args: dict = field(default_factory=lambda: dict(beta=10.0))
+
+    # Steps
+    # ----------------------------------------------------------------------------------------------------
+    def init_loss(self, hparams: HyperParameters):
+        self.loss_layer = GLVQLoss(
+            margin=hparams.margin,
+            transfer_fn=hparams.transfer_fn,
+            **hparams.transfer_args,
+        )
+
+    def loss(self, comparison_measures, batch, components):
+        target = batch[1]
+        comp_labels = components[1]
+        loss = self.loss_layer(comparison_measures, target, comp_labels)
+        self.log('loss', loss)
+        return loss

prototorch/models/architectures/optimization.py

@@ -0,0 +1,73 @@
+from dataclasses import dataclass, field
+from typing import Type
+
+import torch
+from prototorch.models import BaseYArchitecture
+from torch.nn.parameter import Parameter
+
+
+class SingleLearningRateMixin(BaseYArchitecture):
+    """
+    Single Learning Rate
+
+    All parameters are updated with a single learning rate.
+    """
+
+    # HyperParameters
+    # ----------------------------------------------------------------------------------------------------
+    @dataclass
+    class HyperParameters(BaseYArchitecture.HyperParameters):
+        """
+        lr: The learning rate. Default: 0.1.
+        optimizer: The optimizer to use. Default: torch.optim.Adam.
+        """
+        lr: float = 0.1
+        optimizer: Type[torch.optim.Optimizer] = torch.optim.Adam
+
+    # Hooks
+    # ----------------------------------------------------------------------------------------------------
+    def configure_optimizers(self):
+        return self.hparams.optimizer(self.parameters(),
+                                      lr=self.hparams.lr)  # type: ignore
+
+
+class MultipleLearningRateMixin(BaseYArchitecture):
+    """
+    Multiple Learning Rates
+
+    Define Different Learning Rates for different parameters.
+    """
+
+    # HyperParameters
+    # ----------------------------------------------------------------------------------------------------
+    @dataclass
+    class HyperParameters(BaseYArchitecture.HyperParameters):
+        """
+        lr: The learning rate. Default: 0.1.
+        optimizer: The optimizer to use. Default: torch.optim.Adam.
+        """
+        lr: dict = field(default_factory=dict)
+        optimizer: Type[torch.optim.Optimizer] = torch.optim.Adam
+
+    # Hooks
+    # ----------------------------------------------------------------------------------------------------
+    def configure_optimizers(self):
+        optimizers = []
+        for name, lr in self.hparams.lr.items():
+            if not hasattr(self, name):
+                raise ValueError(f"{name} is not a parameter of {self}")
+            else:
+                model_part = getattr(self, name)
+                if isinstance(model_part, Parameter):
+                    optimizers.append(
+                        self.hparams.optimizer(
+                            [model_part],
+                            lr=lr,  # type: ignore
+                        ))
+                elif hasattr(model_part, "parameters"):
+                    optimizers.append(
+                        self.hparams.optimizer(
+                            model_part.parameters(),
+                            lr=lr,  # type: ignore
+                        ))
+        return optimizers

prototorch/models/callbacks.py

@@ -0,0 +1,307 @@
+import logging
+import warnings
+from enum import Enum
+from typing import Optional, Type
+
+import matplotlib.pyplot as plt
+import numpy as np
+import pytorch_lightning as pl
+import torch
+import torchmetrics
+from prototorch.models.architectures.base import BaseYArchitecture, Steps
+from prototorch.models.architectures.comparison import OmegaComparisonMixin
+from prototorch.models.library.gmlvq import GMLVQ
+from prototorch.models.vis import Vis2DAbstract
+from prototorch.utils.utils import mesh2d
+from pytorch_lightning.loggers import TensorBoardLogger
+
+DIVERGING_COLOR_MAPS = [
+    'PiYG',
+    'PRGn',
+    'BrBG',
+    'PuOr',
+    'RdGy',
+    'RdBu',
+    'RdYlBu',
+    'RdYlGn',
+    'Spectral',
+    'coolwarm',
+    'bwr',
+    'seismic',
+]
+
+
+class LogTorchmetricCallback(pl.Callback):
+
+    def __init__(
+        self,
+        name,
+        metric: Type[torchmetrics.Metric],
+        step: str = Steps.TRAINING,
+        on_epoch=True,
+        **metric_kwargs,
+    ) -> None:
+        self.name = name
+        self.metric = metric
+        self.metric_kwargs = metric_kwargs
+        self.step = step
+        self.on_epoch = on_epoch
+
+    def setup(
+        self,
+        trainer: pl.Trainer,
+        pl_module: BaseYArchitecture,
+        stage: Optional[str] = None,
+    ) -> None:
+        pl_module.register_torchmetric(
+            self,
+            self.metric,
+            step=self.step,
+            **self.metric_kwargs,
+        )
+
+    def __call__(self, value, pl_module: BaseYArchitecture):
+        pl_module.log(
+            self.name,
+            value,
+            on_epoch=self.on_epoch,
+            on_step=(not self.on_epoch),
+        )
+
+
+class LogConfusionMatrix(LogTorchmetricCallback):
+
+    def __init__(
+        self,
+        num_classes,
+        name="confusion",
+        on='prediction',
+        **kwargs,
+    ):
+        super().__init__(
+            name,
+            torchmetrics.ConfusionMatrix,
+            on=on,
+            num_classes=num_classes,
+            **kwargs,
+        )
+
+    def __call__(self, value, pl_module: BaseYArchitecture):
+        fig, ax = plt.subplots()
+        ax.imshow(value.detach().cpu().numpy())
+
+        # Show all ticks and label them with the respective list entries
+        # ax.set_xticks(np.arange(len(farmers)), labels=farmers)
+        # ax.set_yticks(np.arange(len(vegetables)), labels=vegetables)
+
+        # Rotate the tick labels and set their alignment.
+        plt.setp(
+            ax.get_xticklabels(),
+            rotation=45,
+            ha="right",
+            rotation_mode="anchor",
+        )
+
+        # Loop over data dimensions and create text annotations.
+        for i in range(len(value)):
+            for j in range(len(value)):
+                text = ax.text(
+                    j,
+                    i,
+                    value[i, j].item(),
+                    ha="center",
+                    va="center",
+                    color="w",
+                )
+
+        ax.set_title(self.name)
+        fig.tight_layout()
+
+        pl_module.logger.experiment.add_figure(
+            tag=self.name,
+            figure=fig,
+            close=True,
+            global_step=pl_module.global_step,
+        )
+
+
+class VisGLVQ2D(Vis2DAbstract):
+
+    def visualize(self, pl_module):
+        protos = pl_module.prototypes
+        plabels = pl_module.prototype_labels
+        x_train, y_train = self.x_train, self.y_train
+        ax = self.setup_ax()
+        self.plot_protos(ax, protos, plabels)
+        if x_train is not None:
+            self.plot_data(ax, x_train, y_train)
+            mesh_input, xx, yy = mesh2d(
+                np.vstack([x_train, protos]),
+                self.border,
+                self.resolution,
+            )
+        else:
+            mesh_input, xx, yy = mesh2d(protos, self.border, self.resolution)
+        _components = pl_module.components_layer.components
+        mesh_input = torch.from_numpy(mesh_input).type_as(_components)
+        y_pred = pl_module.predict(mesh_input)
+        y_pred = y_pred.cpu().reshape(xx.shape)
+        ax.contourf(xx, yy, y_pred, cmap=self.cmap, alpha=0.35)
+
+
+class VisGMLVQ2D(Vis2DAbstract):
+
+    def __init__(self, *args, ev_proj=True, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.ev_proj = ev_proj
+
+    def visualize(self, pl_module):
+        protos = pl_module.prototypes
+        plabels = pl_module.prototype_labels
+        x_train, y_train = self.x_train, self.y_train
+        device = pl_module.device
+        omega = pl_module._omega.detach()
+        lam = omega @ omega.T
+        u, _, _ = torch.pca_lowrank(lam, q=2)
+        with torch.no_grad():
+            x_train = torch.Tensor(x_train).to(device)
+            x_train = x_train @ u
+            x_train = x_train.cpu().detach()
+        if self.show_protos:
+            with torch.no_grad():
+                protos = torch.Tensor(protos).to(device)
+                protos = protos @ u
+                protos = protos.cpu().detach()
+        ax = self.setup_ax()
+        self.plot_data(ax, x_train, y_train)
+        if self.show_protos:
+            self.plot_protos(ax, protos, plabels)
+
+
+class PlotLambdaMatrixToTensorboard(pl.Callback):
+
+    def __init__(self, cmap='seismic') -> None:
+        super().__init__()
+        self.cmap = cmap
+
+        if self.cmap not in DIVERGING_COLOR_MAPS and type(self.cmap) is str:
+            warnings.warn(
+                f"{self.cmap} is not a diverging color map. We recommend to use one of the following: {DIVERGING_COLOR_MAPS}"
+            )
+
+    def on_train_start(self, trainer, pl_module: GMLVQ):
+        self.plot_lambda(trainer, pl_module)
+
+    def on_train_epoch_end(self, trainer, pl_module: GMLVQ):
+        self.plot_lambda(trainer, pl_module)
+
+    def plot_lambda(self, trainer, pl_module: GMLVQ):
+
+        self.fig, self.ax = plt.subplots(1, 1)
+
+        # plot lambda matrix
+        l_matrix = pl_module.lambda_matrix
+
+        # normalize lambda matrix
+        l_matrix = l_matrix / torch.max(torch.abs(l_matrix))
+
+        # plot lambda matrix
+        self.ax.imshow(l_matrix.detach().numpy(), self.cmap, vmin=-1, vmax=1)
+
+        self.fig.colorbar(self.ax.images[-1])
+
+        # add title
+        self.ax.set_title('Lambda Matrix')
+
+        # add to tensorboard
+        if isinstance(trainer.logger, TensorBoardLogger):
+            trainer.logger.experiment.add_figure(
+                "lambda_matrix",
+                self.fig,
+                trainer.global_step,
+            )
+        else:
+            warnings.warn(
+                f"{self.__class__.__name__} is not compatible with {trainer.logger.__class__.__name__} as logger. Use TensorBoardLogger instead."
+            )
+
+
+class Profiles(Enum):
+    '''
+    Available Profiles
+    '''
+    RELEVANCE = 'relevance'
+    INFLUENCE = 'influence'
+
+    def __str__(self):
+        return str(self.value)
+
+
+class PlotMatrixProfiles(pl.Callback):
+
+    def __init__(self, profile=Profiles.INFLUENCE, cmap='seismic') -> None:
+        super().__init__()
+        self.cmap = cmap
+        self.profile = profile
+
+    def on_train_start(self, trainer, pl_module: GMLVQ):
+        '''
+        Plot initial profile.
+        '''
+        self._plot_profile(trainer, pl_module)
+
+    def on_train_epoch_end(self, trainer, pl_module: GMLVQ):
+        '''
+        Plot after every epoch.
+        '''
+        self._plot_profile(trainer, pl_module)
+
+    def _plot_profile(self, trainer, pl_module: GMLVQ):
+
+        fig, ax = plt.subplots(1, 1)
+
+        # plot lambda matrix
+        l_matrix = torch.abs(pl_module.lambda_matrix)
+
+        if self.profile == Profiles.RELEVANCE:
+            profile_value = l_matrix.diag()
+        elif self.profile == Profiles.INFLUENCE:
+            profile_value = l_matrix.sum(0)
+
+        # plot lambda matrix
+        ax.plot(profile_value.detach().numpy())
+
+        # add title
+        ax.set_title(f'{self.profile} profile')
+
+        # add to tensorboard
+        if isinstance(trainer.logger, TensorBoardLogger):
+            trainer.logger.experiment.add_figure(
+                f"{self.profile}_matrix",
+                fig,
+                trainer.global_step,
+            )
+        else:
+            class_name = self.__class__.__name__
+            logger_name = trainer.logger.__class__.__name__
+            warnings.warn(
+                f"{class_name} is not compatible with {logger_name} as logger. Use TensorBoardLogger instead."
+            )
+
+
+class OmegaTraceNormalization(pl.Callback):
+    '''
+    Trace normalization of the Omega Matrix.
+    '''
+    __epsilon = torch.finfo(torch.float32).eps
+
+    def on_train_epoch_end(self, trainer: "pl.Trainer",
+                           pl_module: OmegaComparisonMixin) -> None:
+
+        omega = pl_module.parameter_omega
+        denominator = torch.sqrt(torch.trace(omega.T @ omega))
+        logging.debug(
+            "Apply Omega Trace Normalization: demoninator=%f",
+            denominator.item(),
+        )
+        pl_module.parameter_omega = omega / (denominator + self.__epsilon)

prototorch/models/library/__init__.py

@@ -0,0 +1,7 @@
+from .glvq import GLVQ
+from .gmlvq import GMLVQ
+
+__all__ = [
+    "GLVQ",
+    "GMLVQ",
+]

prototorch/models/library/glvq.py

@@ -0,0 +1,35 @@
+from dataclasses import dataclass
+
+from prototorch.models import (
+    SimpleComparisonMixin,
+    SingleLearningRateMixin,
+    SupervisedArchitecture,
+    WTACompetitionMixin,
+)
+from prototorch.models.architectures.loss import GLVQLossMixin
+
+
+class GLVQ(
+        SupervisedArchitecture,
+        SimpleComparisonMixin,
+        GLVQLossMixin,
+        WTACompetitionMixin,
+        SingleLearningRateMixin,
+):
+    """
+    Generalized Learning Vector Quantization (GLVQ)
+
+    A GLVQ architecture that uses the winner-take-all strategy and the GLVQ loss.
+    """
+
+    @dataclass
+    class HyperParameters(
+            SimpleComparisonMixin.HyperParameters,
+            SingleLearningRateMixin.HyperParameters,
+            GLVQLossMixin.HyperParameters,
+            WTACompetitionMixin.HyperParameters,
+            SupervisedArchitecture.HyperParameters,
+    ):
+        """
+        No hyperparameters.
+        """

prototorch/models/library/gmlvq.py

@@ -0,0 +1,50 @@
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+from typing import Callable
+
+import torch
+from prototorch.core.distances import omega_distance
+from prototorch.models import (
+    GLVQLossMixin,
+    MultipleLearningRateMixin,
+    OmegaComparisonMixin,
+    SupervisedArchitecture,
+    WTACompetitionMixin,
+)
+
+
+class GMLVQ(
+        SupervisedArchitecture,
+        OmegaComparisonMixin,
+        GLVQLossMixin,
+        WTACompetitionMixin,
+        MultipleLearningRateMixin,
+):
+    """
+    Generalized Matrix Learning Vector Quantization (GMLVQ)
+
+    A GMLVQ architecture that uses the winner-take-all strategy and the GLVQ loss.
+    """
+    # HyperParameters
+    # ----------------------------------------------------------------------------------------------------
+    @dataclass
+    class HyperParameters(
+            MultipleLearningRateMixin.HyperParameters,
+            OmegaComparisonMixin.HyperParameters,
+            GLVQLossMixin.HyperParameters,
+            WTACompetitionMixin.HyperParameters,
+            SupervisedArchitecture.HyperParameters,
+    ):
+        """
+        comparison_fn: The comparison / dissimilarity function to use. Override Default: omega_distance.
+        comparison_args: Keyword arguments for the comparison function. Override Default: {}.
+        """
+        comparison_fn: Callable = omega_distance
+        comparison_args: dict = field(default_factory=dict)
+        optimizer: type[torch.optim.Optimizer] = torch.optim.Adam
+
+        lr: dict = field(default_factory=lambda: dict(
+            components_layer=0.1,
+            _omega=0.5,
+        ))

pyproject.toml

@@ -1,90 +0,0 @@
-
-[project]
-name = "prototorch-models"
-version = "0.7.1"
-description = "Pre-packaged prototype-based machine learning models using ProtoTorch and PyTorch-Lightning."
-authors = [
-    { name = "Jensun Ravichandran", email = "jjensun@gmail.com" },
-    { name = "Alexander Engelsberger", email = "engelsbe@hs-mittweida.de" },
-]
-dependencies = ["lightning>=2.0.0", "prototorch>=0.7.5"]
-requires-python = ">=3.8"
-readme = "README.md"
-license = { text = "MIT" }
-classifiers = [
-    "Development Status :: 2 - Pre-Alpha",
-    "Environment :: Plugins",
-    "Intended Audience :: Developers",
-    "Intended Audience :: Education",
-    "Intended Audience :: Science/Research",
-    "License :: OSI Approved :: MIT License",
-    "Natural Language :: English",
-    "Operating System :: OS Independent",
-    "Programming Language :: Python :: 3",
-    "Programming Language :: Python :: 3.10",
-    "Programming Language :: Python :: 3.11",
-    "Programming Language :: Python :: 3.8",
-    "Programming Language :: Python :: 3.9",
-    "Topic :: Scientific/Engineering :: Artificial Intelligence",
-    "Topic :: Software Development :: Libraries",
-    "Topic :: Software Development :: Libraries :: Python Modules",
-]
-
-[project.urls]
-Homepage = "https://github.com/si-cim/prototorch_models"
-Downloads = "https://github.com/si-cim/prototorch_models.git"
-
-[project.optional-dependencies]
-dev = ["bumpversion", "pre-commit", "yapf", "toml"]
-examples = ["matplotlib", "scikit-learn"]
-ci = ["pytest", "pre-commit"]
-docs = [
-    "recommonmark",
-    "nbsphinx",
-    "sphinx",
-    "sphinx_rtd_theme",
-    "sphinxcontrib-bibtex",
-    "sphinxcontrib-katex",
-    "ipykernel",
-]
-all = [
-    "bumpversion",
-    "pre-commit",
-    "yapf",
-    "toml",
-    "pytest",
-    "matplotlib",
-    "scikit-learn",
-    "recommonmark",
-    "nbsphinx",
-    "sphinx",
-    "sphinx_rtd_theme",
-    "sphinxcontrib-bibtex",
-    "sphinxcontrib-katex",
-    "ipykernel",
-]
-
-[build-system]
-requires = ["setuptools>=61", "wheel"]
-build-backend = "setuptools.build_meta"
-
-[tool.yapf]
-based_on_style = "pep8"
-spaces_before_comment = 2
-split_before_logical_operator = true
-
-[tool.pylint]
-disable = ["too-many-arguments", "too-few-public-methods", "fixme"]
-
-[tool.isort]
-profile = "hug"
-src_paths = ["isort", "test"]
-multi_line_output = 3
-include_trailing_comma = true
-force_grid_wrap = 3
-use_parentheses = true
-line_length = 79
-
-[tool.mypy]
-explicit_package_bases = true
-namespace_packages = true

setup.cfg

@@ -0,0 +1,23 @@
+[yapf]
+based_on_style = pep8
+spaces_before_comment = 2
+split_before_logical_operator = true
+
+[pylint]
+disable =
+	too-many-arguments,
+	too-few-public-methods,
+	fixme,
+
+
+[pycodestyle]
+max-line-length = 79
+
+[isort]
+profile = hug
+src_paths = isort, test
+multi_line_output = 3
+include_trailing_comma = True
+force_grid_wrap = 3
+use_parentheses = True
+line_length = 79

setup.py

@@ -0,0 +1,100 @@
+"""
+
+ ######
+ #     # #####   ####  #####  ####  #####  ####  #####   ####  #    #
+ #     # #    # #    #   #   #    #   #   #    # #    # #    # #    #
+ ######  #    # #    #   #   #    #   #   #    # #    # #      ######
+ #       #####  #    #   #   #    #   #   #    # #####  #      #    #
+ #       #   #  #    #   #   #    #   #   #    # #   #  #    # #    #
+ #       #    #  ####    #    ####    #    ####  #    #  ####  #    #Plugin
+
+ProtoTorch models Plugin Package
+"""
+from pathlib import Path
+
+from pkg_resources import safe_name
+from setuptools import find_namespace_packages, setup
+
+PLUGIN_NAME = "models"
+
+PROJECT_URL = "https://github.com/si-cim/prototorch_models"
+DOWNLOAD_URL = "https://github.com/si-cim/prototorch_models.git"
+
+long_description = Path("README.md").read_text(encoding='utf8')
+
+INSTALL_REQUIRES = [
+    "prototorch>=0.7.3",
+    "pytorch_lightning>=1.6.0",
+    "torchmetrics",
+    "protobuf<3.20.0",
+]
+CLI = [
+    "jsonargparse",
+]
+DEV = [
+    "bumpversion",
+    "pre-commit",
+]
+DOCS = [
+    "recommonmark",
+    "sphinx",
+    "nbsphinx",
+    "ipykernel",
+    "sphinx_rtd_theme",
+    "sphinxcontrib-katex",
+    "sphinxcontrib-bibtex",
+]
+EXAMPLES = [
+    "matplotlib",
+    "scikit-learn",
+]
+TESTS = [
+    "codecov",
+    "pytest",
+]
+ALL = CLI + DEV + DOCS + EXAMPLES + TESTS
+
+setup(
+    name=safe_name("prototorch_" + PLUGIN_NAME),
+    version="1.0.0-a8",
+    description="Pre-packaged prototype-based "
+    "machine learning models using ProtoTorch and PyTorch-Lightning.",
+    long_description=long_description,
+    long_description_content_type="text/markdown",
+    author="Alexander Engelsberger",
+    author_email="engelsbe@hs-mittweida.de",
+    url=PROJECT_URL,
+    download_url=DOWNLOAD_URL,
+    license="MIT",
+    python_requires=">=3.7",
+    install_requires=INSTALL_REQUIRES,
+    extras_require={
+        "dev": DEV,
+        "examples": EXAMPLES,
+        "tests": TESTS,
+        "all": ALL,
+    },
+    classifiers=[
+        "Development Status :: 2 - Pre-Alpha",
+        "Environment :: Plugins",
+        "Intended Audience :: Developers",
+        "Intended Audience :: Education",
+        "Intended Audience :: Science/Research",
+        "License :: OSI Approved :: MIT License",
+        "Natural Language :: English",
+        "Programming Language :: Python :: 3",
+        "Programming Language :: Python :: 3.10",
+        "Programming Language :: Python :: 3.9",
+        "Programming Language :: Python :: 3.8",
+        "Programming Language :: Python :: 3.7",
+        "Operating System :: OS Independent",
+        "Topic :: Scientific/Engineering :: Artificial Intelligence",
+        "Topic :: Software Development :: Libraries",
+        "Topic :: Software Development :: Libraries :: Python Modules",
+    ],
+    entry_points={
+        "prototorch.plugins": f"{PLUGIN_NAME} = prototorch.{PLUGIN_NAME}"
+    },
+    packages=find_namespace_packages(include=["prototorch.*"]),
+    zip_safe=False,
+)

src/prototorch/models/__init__.py

@@ -1,39 +0,0 @@
-"""`models` plugin for the `prototorch` package."""
-
-from .callbacks import PrototypeConvergence, PruneLoserPrototypes
-from .cbc import CBC, ImageCBC
-from .glvq import (
-    GLVQ,
-    GLVQ1,
-    GLVQ21,
-    GMLVQ,
-    GRLVQ,
-    GTLVQ,
-    LGMLVQ,
-    LVQMLN,
-    ImageGLVQ,
-    ImageGMLVQ,
-    ImageGTLVQ,
-    SiameseGLVQ,
-    SiameseGMLVQ,
-    SiameseGTLVQ,
-)
-from .knn import KNN
-from .lvq import (
-    LVQ1,
-    LVQ21,
-    MedianLVQ,
-)
-from .probabilistic import (
-    CELVQ,
-    RSLVQ,
-    SLVQ,
-)
-from .unsupervised import (
-    GrowingNeuralGas,
-    KohonenSOM,
-    NeuralGas,
-)
-from .vis import *
-
-__version__ = "0.7.1"

src/prototorch/models/abstract.py

@@ -1,248 +0,0 @@
-"""Abstract classes to be inherited by prototorch models."""
-
-import logging
-
-import pytorch_lightning as pl
-import torch
-import torch.nn.functional as F
-import torchmetrics
-from prototorch.core.competitions import WTAC
-from prototorch.core.components import (
-    AbstractComponents,
-    Components,
-    LabeledComponents,
-)
-from prototorch.core.distances import euclidean_distance
-from prototorch.core.initializers import (
-    LabelsInitializer,
-    ZerosCompInitializer,
-)
-from prototorch.core.pooling import stratified_min_pooling
-from prototorch.nn.wrappers import LambdaLayer
-
-
-class ProtoTorchBolt(pl.LightningModule):
-    """All ProtoTorch models are ProtoTorch Bolts."""
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__()
-
-        # Hyperparameters
-        self.save_hyperparameters(hparams)
-
-        # Default hparams
-        self.hparams.setdefault("lr", 0.01)
-
-        # Default config
-        self.optimizer = kwargs.get("optimizer", torch.optim.Adam)
-        self.lr_scheduler = kwargs.get("lr_scheduler", None)
-        self.lr_scheduler_kwargs = kwargs.get("lr_scheduler_kwargs", dict())
-
-    def configure_optimizers(self):
-        optimizer = self.optimizer(self.parameters(), lr=self.hparams["lr"])
-        if self.lr_scheduler is not None:
-            scheduler = self.lr_scheduler(optimizer,
-                                          **self.lr_scheduler_kwargs)
-            sch = {
-                "scheduler": scheduler,
-                "interval": "step",
-            }  # called after each training step
-            return [optimizer], [sch]
-        else:
-            return optimizer
-
-    def reconfigure_optimizers(self):
-        if self.trainer:
-            self.trainer.strategy.setup_optimizers(self.trainer)
-        else:
-            logging.warning("No trainer to reconfigure optimizers!")
-
-    def __repr__(self):
-        surep = super().__repr__()
-        indented = "".join([f"\t{line}\n" for line in surep.splitlines()])
-        wrapped = f"ProtoTorch Bolt(\n{indented})"
-        return wrapped
-
-
-class PrototypeModel(ProtoTorchBolt):
-    proto_layer: AbstractComponents
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        distance_fn = kwargs.get("distance_fn", euclidean_distance)
-        self.distance_layer = LambdaLayer(distance_fn, name="distance_fn")
-
-    @property
-    def num_prototypes(self):
-        return len(self.proto_layer.components)
-
-    @property
-    def prototypes(self):
-        return self.proto_layer.components.detach().cpu()
-
-    @property
-    def components(self):
-        """Only an alias for the prototypes."""
-        return self.prototypes
-
-    def add_prototypes(self, *args, **kwargs):
-        self.proto_layer.add_components(*args, **kwargs)
-        self.hparams["distribution"] = self.proto_layer.distribution
-        self.reconfigure_optimizers()
-
-    def remove_prototypes(self, indices):
-        self.proto_layer.remove_components(indices)
-        self.hparams["distribution"] = self.proto_layer.distribution
-        self.reconfigure_optimizers()
-
-
-class UnsupervisedPrototypeModel(PrototypeModel):
-    proto_layer: Components
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        # Layers
-        prototypes_initializer = kwargs.get("prototypes_initializer", None)
-        if prototypes_initializer is not None:
-            self.proto_layer = Components(
-                self.hparams["num_prototypes"],
-                initializer=prototypes_initializer,
-            )
-
-    def compute_distances(self, x):
-        protos = self.proto_layer().type_as(x)
-        distances = self.distance_layer(x, protos)
-        return distances
-
-    def forward(self, x):
-        distances = self.compute_distances(x)
-        return distances
-
-
-class SupervisedPrototypeModel(PrototypeModel):
-    proto_layer: LabeledComponents
-
-    def __init__(self, hparams, skip_proto_layer=False, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        # Layers
-        distribution = hparams.get("distribution", None)
-        prototypes_initializer = kwargs.get("prototypes_initializer", None)
-        labels_initializer = kwargs.get("labels_initializer",
-                                        LabelsInitializer())
-        if not skip_proto_layer:
-            # when subclasses do not need a customized prototype layer
-            if prototypes_initializer is not None:
-                # when building a new model
-                self.proto_layer = LabeledComponents(
-                    distribution=distribution,
-                    components_initializer=prototypes_initializer,
-                    labels_initializer=labels_initializer,
-                )
-                proto_shape = self.proto_layer.components.shape[1:]
-                self.hparams["initialized_proto_shape"] = proto_shape
-            else:
-                # when restoring a checkpointed model
-                self.proto_layer = LabeledComponents(
-                    distribution=distribution,
-                    components_initializer=ZerosCompInitializer(
-                        self.hparams["initialized_proto_shape"]),
-                )
-        self.competition_layer = WTAC()
-
-    @property
-    def prototype_labels(self):
-        return self.proto_layer.labels.detach().cpu()
-
-    @property
-    def num_classes(self):
-        return self.proto_layer.num_classes
-
-    def compute_distances(self, x):
-        protos, _ = self.proto_layer()
-        distances = self.distance_layer(x, protos)
-        return distances
-
-    def forward(self, x):
-        distances = self.compute_distances(x)
-        _, plabels = self.proto_layer()
-        winning = stratified_min_pooling(distances, plabels)
-        y_pred = F.softmin(winning, dim=1)
-        return y_pred
-
-    def predict_from_distances(self, distances):
-        with torch.no_grad():
-            _, plabels = self.proto_layer()
-            y_pred = self.competition_layer(distances, plabels)
-        return y_pred
-
-    def predict(self, x):
-        with torch.no_grad():
-            distances = self.compute_distances(x)
-        y_pred = self.predict_from_distances(distances)
-        return y_pred
-
-    def log_acc(self, distances, targets, tag):
-        preds = self.predict_from_distances(distances)
-        accuracy = torchmetrics.functional.accuracy(
-            preds.int(),
-            targets.int(),
-            "multiclass",
-            num_classes=self.num_classes,
-        )
-
-        self.log(
-            tag,
-            accuracy,
-            on_step=False,
-            on_epoch=True,
-            prog_bar=True,
-            logger=True,
-        )
-
-    def test_step(self, batch, batch_idx):
-        x, targets = batch
-
-        preds = self.predict(x)
-        accuracy = torchmetrics.functional.accuracy(
-            preds.int(),
-            targets.int(),
-            "multiclass",
-            num_classes=self.num_classes,
-        )
-
-        self.log("test_acc", accuracy)
-
-
-class ProtoTorchMixin:
-    """All mixins are ProtoTorchMixins."""
-
-
-class NonGradientMixin(ProtoTorchMixin):
-    """Mixin for custom non-gradient optimization."""
-
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-        self.automatic_optimization = False
-
-    def training_step(self, train_batch, batch_idx):
-        raise NotImplementedError
-
-
-class ImagePrototypesMixin(ProtoTorchMixin):
-    """Mixin for models with image prototypes."""
-    proto_layer: Components
-    components: torch.Tensor
-
-    def on_train_batch_end(self, outputs, batch, batch_idx):
-        """Constrain the components to the range [0, 1] by clamping after updates."""
-        self.proto_layer.components.data.clamp_(0.0, 1.0)
-
-    def get_prototype_grid(self, num_columns=2, return_channels_last=True):
-        from torchvision.utils import make_grid
-        grid = make_grid(self.components, nrow=num_columns)
-        if return_channels_last:
-            grid = grid.permute((1, 2, 0))
-        return grid.cpu()

src/prototorch/models/callbacks.py

@@ -1,152 +0,0 @@
-"""Lightning Callbacks."""
-
-import logging
-from typing import TYPE_CHECKING
-
-import pytorch_lightning as pl
-import torch
-from prototorch.core.initializers import LiteralCompInitializer
-
-from .extras import ConnectionTopology
-
-if TYPE_CHECKING:
-    from prototorch.models import GLVQ, GrowingNeuralGas
-
-
-class PruneLoserPrototypes(pl.Callback):
-
-    def __init__(
-        self,
-        threshold=0.01,
-        idle_epochs=10,
-        prune_quota_per_epoch=-1,
-        frequency=1,
-        replace=False,
-        prototypes_initializer=None,
-        verbose=False,
-    ):
-        self.threshold = threshold  # minimum win ratio
-        self.idle_epochs = idle_epochs  # epochs to wait before pruning
-        self.prune_quota_per_epoch = prune_quota_per_epoch
-        self.frequency = frequency
-        self.replace = replace
-        self.verbose = verbose
-        self.prototypes_initializer = prototypes_initializer
-
-    def on_train_epoch_end(self, trainer, pl_module: "GLVQ"):
-        if (trainer.current_epoch + 1) < self.idle_epochs:
-            return None
-        if (trainer.current_epoch + 1) % self.frequency:
-            return None
-
-        ratios = pl_module.prototype_win_ratios.mean(dim=0)
-        to_prune = torch.arange(len(ratios))[ratios < self.threshold]
-        to_prune = to_prune.tolist()
-        prune_labels = pl_module.prototype_labels[to_prune]
-        if self.prune_quota_per_epoch > 0:
-            to_prune = to_prune[:self.prune_quota_per_epoch]
-            prune_labels = prune_labels[:self.prune_quota_per_epoch]
-
-        if len(to_prune) > 0:
-            logging.debug(f"\nPrototype win ratios: {ratios}")
-            logging.debug(f"Pruning prototypes at: {to_prune}")
-            logging.debug(f"Corresponding labels are: {prune_labels.tolist()}")
-
-            cur_num_protos = pl_module.num_prototypes
-            pl_module.remove_prototypes(indices=to_prune)
-
-            if self.replace:
-                labels, counts = torch.unique(prune_labels,
-                                              sorted=True,
-                                              return_counts=True)
-                distribution = dict(zip(labels.tolist(), counts.tolist()))
-
-                logging.info(f"Re-adding pruned prototypes...")
-                logging.debug(f"distribution={distribution}")
-
-                pl_module.add_prototypes(
-                    distribution=distribution,
-                    components_initializer=self.prototypes_initializer)
-            new_num_protos = pl_module.num_prototypes
-
-            logging.info(f"`num_prototypes` changed from {cur_num_protos} "
-                         f"to {new_num_protos}.")
-        return True
-
-
-class PrototypeConvergence(pl.Callback):
-
-    def __init__(self, min_delta=0.01, idle_epochs=10, verbose=False):
-        self.min_delta = min_delta
-        self.idle_epochs = idle_epochs  # epochs to wait
-        self.verbose = verbose
-
-    def on_train_epoch_end(self, trainer, pl_module):
-        if (trainer.current_epoch + 1) < self.idle_epochs:
-            return None
-
-        logging.info("Stopping...")
-        # TODO
-        return True
-
-
-class GNGCallback(pl.Callback):
-    """GNG Callback.
-
-    Applies growing algorithm based on accumulated error and topology.
-
-    Based on "A Growing Neural Gas Network Learns Topologies" by Bernd Fritzke.
-
-    """
-
-    def __init__(self, reduction=0.1, freq=10):
-        self.reduction = reduction
-        self.freq = freq
-
-    def on_train_epoch_end(
-        self,
-        trainer: pl.Trainer,
-        pl_module: "GrowingNeuralGas",
-    ):
-        if (trainer.current_epoch + 1) % self.freq == 0:
-            # Get information
-            errors = pl_module.errors
-            topology: ConnectionTopology = pl_module.topology_layer
-            components = pl_module.proto_layer.components
-
-            # Insertion point
-            worst = torch.argmax(errors)
-
-            neighbors = topology.get_neighbors(worst)[0]
-
-            if len(neighbors) == 0:
-                logging.log(level=20, msg="No neighbor-pairs found!")
-                return
-
-            neighbors_errors = errors[neighbors]
-            worst_neighbor = neighbors[torch.argmax(neighbors_errors)]
-
-            # New Prototype
-            new_component = 0.5 * (components[worst] +
-                                   components[worst_neighbor])
-
-            # Add component
-            pl_module.proto_layer.add_components(
-                1,
-                initializer=LiteralCompInitializer(new_component.unsqueeze(0)),
-            )
-
-            # Adjust Topology
-            topology.add_prototype()
-            topology.add_connection(worst, -1)
-            topology.add_connection(worst_neighbor, -1)
-            topology.remove_connection(worst, worst_neighbor)
-
-            # New errors
-            worst_error = errors[worst].unsqueeze(0)
-            pl_module.errors = torch.cat([pl_module.errors, worst_error])
-            pl_module.errors[worst] = errors[worst] * self.reduction
-            pl_module.errors[
-                worst_neighbor] = errors[worst_neighbor] * self.reduction
-
-            trainer.strategy.setup_optimizers(trainer)

src/prototorch/models/cbc.py

@@ -1,84 +0,0 @@
-import torch
-import torchmetrics
-from prototorch.core.competitions import CBCC
-from prototorch.core.components import ReasoningComponents
-from prototorch.core.initializers import RandomReasoningsInitializer
-from prototorch.core.losses import MarginLoss
-from prototorch.core.similarities import euclidean_similarity
-from prototorch.nn.wrappers import LambdaLayer
-
-from .abstract import ImagePrototypesMixin
-from .glvq import SiameseGLVQ
-
-
-class CBC(SiameseGLVQ):
-    """Classification-By-Components."""
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, skip_proto_layer=True, **kwargs)
-
-        similarity_fn = kwargs.get("similarity_fn", euclidean_similarity)
-        components_initializer = kwargs.get("components_initializer", None)
-        reasonings_initializer = kwargs.get("reasonings_initializer",
-                                            RandomReasoningsInitializer())
-        self.components_layer = ReasoningComponents(
-            self.hparams.distribution,
-            components_initializer=components_initializer,
-            reasonings_initializer=reasonings_initializer,
-        )
-        self.similarity_layer = LambdaLayer(similarity_fn)
-        self.competition_layer = CBCC()
-
-        # Namespace hook
-        self.proto_layer = self.components_layer
-
-        self.loss = MarginLoss(self.hparams.margin)
-
-    def forward(self, x):
-        components, reasonings = self.components_layer()
-        latent_x = self.backbone(x)
-        self.backbone.requires_grad_(self.both_path_gradients)
-        latent_components = self.backbone(components)
-        self.backbone.requires_grad_(True)
-        detections = self.similarity_layer(latent_x, latent_components)
-        probs = self.competition_layer(detections, reasonings)
-        return probs
-
-    def shared_step(self, batch, batch_idx):
-        x, y = batch
-        y_pred = self(x)
-        num_classes = self.num_classes
-        y_true = torch.nn.functional.one_hot(y.long(), num_classes=num_classes)
-        loss = self.loss(y_pred, y_true).mean()
-        return y_pred, loss
-
-    def training_step(self, batch, batch_idx):
-        y_pred, train_loss = self.shared_step(batch, batch_idx)
-        preds = torch.argmax(y_pred, dim=1)
-        accuracy = torchmetrics.functional.accuracy(
-            preds.int(),
-            batch[1].int(),
-            "multiclass",
-            num_classes=self.num_classes,
-        )
-        self.log(
-            "train_acc",
-            accuracy,
-            on_step=False,
-            on_epoch=True,
-            prog_bar=True,
-            logger=True,
-        )
-        return train_loss
-
-    def predict(self, x):
-        with torch.no_grad():
-            y_pred = self(x)
-            y_pred = torch.argmax(y_pred, dim=1)
-        return y_pred
-
-
-class ImageCBC(ImagePrototypesMixin, CBC):
-    """CBC model that constrains the components to the range [0, 1] by
-    clamping after updates.
-    """

src/prototorch/models/extras.py

@@ -1,130 +0,0 @@
-"""prototorch.models.extras
-
-Modules not yet available in prototorch go here temporarily.
-
-"""
-
-import torch
-from prototorch.core.similarities import gaussian
-
-
-def rank_scaled_gaussian(distances, lambd):
-    order = torch.argsort(distances, dim=1)
-    ranks = torch.argsort(order, dim=1)
-    return torch.exp(-torch.exp(-ranks / lambd) * distances)
-
-
-def orthogonalization(tensors):
-    """Orthogonalization via polar decomposition """
-    u, _, v = torch.svd(tensors, compute_uv=True)
-    u_shape = tuple(list(u.shape))
-    v_shape = tuple(list(v.shape))
-
-    # reshape to (num x N x M)
-    u = torch.reshape(u, (-1, u_shape[-2], u_shape[-1]))
-    v = torch.reshape(v, (-1, v_shape[-2], v_shape[-1]))
-
-    out = u @ v.permute([0, 2, 1])
-
-    out = torch.reshape(out, u_shape[:-1] + (v_shape[-2], ))
-
-    return out
-
-
-def ltangent_distance(x, y, omegas):
-    r"""Localized Tangent distance.
-    Compute Orthogonal Complement: math:`\bm P_k = \bm I - \Omega_k \Omega_k^T`
-    Compute Tangent Distance: math:`{\| \bm P \bm x - \bm P_k \bm y_k \|}_2`
-
-    :param `torch.tensor` omegas: Three dimensional matrix
-    :rtype: `torch.tensor`
-    """
-    x, y = (arr.view(arr.size(0), -1) for arr in (x, y))
-    p = torch.eye(omegas.shape[-2], device=omegas.device) - torch.bmm(
-        omegas, omegas.permute([0, 2, 1]))
-    projected_x = x @ p
-    projected_y = torch.diagonal(y @ p).T
-    expanded_y = torch.unsqueeze(projected_y, dim=1)
-    batchwise_difference = expanded_y - projected_x
-    differences_squared = batchwise_difference**2
-    distances = torch.sqrt(torch.sum(differences_squared, dim=2))
-    distances = distances.permute(1, 0)
-    return distances
-
-
-class GaussianPrior(torch.nn.Module):
-
-    def __init__(self, variance):
-        super().__init__()
-        self.variance = variance
-
-    def forward(self, distances):
-        return gaussian(distances, self.variance)
-
-
-class RankScaledGaussianPrior(torch.nn.Module):
-
-    def __init__(self, lambd):
-        super().__init__()
-        self.lambd = lambd
-
-    def forward(self, distances):
-        return rank_scaled_gaussian(distances, self.lambd)
-
-
-class ConnectionTopology(torch.nn.Module):
-
-    def __init__(self, agelimit, num_prototypes):
-        super().__init__()
-        self.agelimit = agelimit
-        self.num_prototypes = num_prototypes
-
-        self.cmat = torch.zeros((self.num_prototypes, self.num_prototypes))
-        self.age = torch.zeros_like(self.cmat)
-
-    def forward(self, d):
-        order = torch.argsort(d, dim=1)
-
-        for element in order:
-            i0, i1 = element[0], element[1]
-
-            self.cmat[i0][i1] = 1
-            self.cmat[i1][i0] = 1
-
-            self.age[i0][i1] = 0
-            self.age[i1][i0] = 0
-
-            self.age[i0][self.cmat[i0] == 1] += 1
-            self.age[i1][self.cmat[i1] == 1] += 1
-
-            self.cmat[i0][self.age[i0] > self.agelimit] = 0
-            self.cmat[i1][self.age[i1] > self.agelimit] = 0
-
-    def get_neighbors(self, position):
-        return torch.where(self.cmat[position])
-
-    def add_prototype(self):
-        new_cmat = torch.zeros([dim + 1 for dim in self.cmat.shape])
-        new_cmat[:-1, :-1] = self.cmat
-        self.cmat = new_cmat
-
-        new_age = torch.zeros([dim + 1 for dim in self.age.shape])
-        new_age[:-1, :-1] = self.age
-        self.age = new_age
-
-    def add_connection(self, a, b):
-        self.cmat[a][b] = 1
-        self.cmat[b][a] = 1
-
-        self.age[a][b] = 0
-        self.age[b][a] = 0
-
-    def remove_connection(self, a, b):
-        self.cmat[a][b] = 0
-        self.cmat[b][a] = 0
-
-        self.age[a][b] = 0
-        self.age[b][a] = 0
-
-    def extra_repr(self):
-        return f"(agelimit): ({self.agelimit})"

src/prototorch/models/glvq.py

@@ -1,381 +0,0 @@
-"""Models based on the GLVQ framework."""
-
-import torch
-from prototorch.core.competitions import wtac
-from prototorch.core.distances import (
-    lomega_distance,
-    omega_distance,
-    squared_euclidean_distance,
-)
-from prototorch.core.initializers import EyeLinearTransformInitializer
-from prototorch.core.losses import (
-    GLVQLoss,
-    lvq1_loss,
-    lvq21_loss,
-)
-from prototorch.core.transforms import LinearTransform
-from prototorch.nn.wrappers import LambdaLayer, LossLayer
-from torch.nn.parameter import Parameter
-
-from .abstract import ImagePrototypesMixin, SupervisedPrototypeModel
-from .extras import ltangent_distance, orthogonalization
-
-
-class GLVQ(SupervisedPrototypeModel):
-    """Generalized Learning Vector Quantization."""
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        # Default hparams
-        self.hparams.setdefault("margin", 0.0)
-        self.hparams.setdefault("transfer_fn", "identity")
-        self.hparams.setdefault("transfer_beta", 10.0)
-
-        # Loss
-        self.loss = GLVQLoss(
-            margin=self.hparams["margin"],
-            transfer_fn=self.hparams["transfer_fn"],
-            beta=self.hparams["transfer_beta"],
-        )
-
-    # def on_save_checkpoint(self, checkpoint):
-    #     if "prototype_win_ratios" in checkpoint["state_dict"]:
-    #         del checkpoint["state_dict"]["prototype_win_ratios"]
-
-    def initialize_prototype_win_ratios(self):
-        self.register_buffer(
-            "prototype_win_ratios",
-            torch.zeros(self.num_prototypes, device=self.device))
-
-    def on_train_epoch_start(self):
-        self.initialize_prototype_win_ratios()
-
-    def log_prototype_win_ratios(self, distances):
-        batch_size = len(distances)
-        prototype_wc = torch.zeros(self.num_prototypes,
-                                   dtype=torch.long,
-                                   device=self.device)
-        wi, wc = torch.unique(distances.min(dim=-1).indices,
-                              sorted=True,
-                              return_counts=True)
-        prototype_wc[wi] = wc
-        prototype_wr = prototype_wc / batch_size
-        self.prototype_win_ratios = torch.vstack([
-            self.prototype_win_ratios,
-            prototype_wr,
-        ])
-
-    def shared_step(self, batch, batch_idx):
-        x, y = batch
-        out = self.compute_distances(x)
-        _, plabels = self.proto_layer()
-        loss = self.loss(out, y, plabels)
-        return out, loss
-
-    def training_step(self, batch, batch_idx):
-        out, train_loss = self.shared_step(batch, batch_idx)
-        self.log_prototype_win_ratios(out)
-        self.log("train_loss", train_loss)
-        self.log_acc(out, batch[-1], tag="train_acc")
-        return train_loss
-
-    def validation_step(self, batch, batch_idx):
-        # `model.eval()` and `torch.no_grad()` handled by pl
-        out, val_loss = self.shared_step(batch, batch_idx)
-        self.log("val_loss", val_loss)
-        self.log_acc(out, batch[-1], tag="val_acc")
-        return val_loss
-
-    def test_step(self, batch, batch_idx):
-        # `model.eval()` and `torch.no_grad()` handled by pl
-        out, test_loss = self.shared_step(batch, batch_idx)
-        self.log_acc(out, batch[-1], tag="test_acc")
-        return test_loss
-
-    def test_epoch_end(self, outputs):
-        test_loss = 0.0
-        for batch_loss in outputs:
-            test_loss += batch_loss.item()
-        self.log("test_loss", test_loss)
-
-
-class SiameseGLVQ(GLVQ):
-    """GLVQ in a Siamese setting.
-
-    GLVQ model that applies an arbitrary transformation on the inputs and the
-    prototypes before computing the distances between them. The weights in the
-    transformation pipeline are only learned from the inputs.
-
-    """
-
-    def __init__(self,
-                 hparams,
-                 backbone=torch.nn.Identity(),
-                 both_path_gradients=False,
-                 **kwargs):
-        distance_fn = kwargs.pop("distance_fn", squared_euclidean_distance)
-        super().__init__(hparams, distance_fn=distance_fn, **kwargs)
-        self.backbone = backbone
-        self.both_path_gradients = both_path_gradients
-
-    def compute_distances(self, x):
-        protos, _ = self.proto_layer()
-        x, protos = (arr.view(arr.size(0), -1) for arr in (x, protos))
-        latent_x = self.backbone(x)
-
-        bb_grad = any([el.requires_grad for el in self.backbone.parameters()])
-
-        self.backbone.requires_grad_(bb_grad and self.both_path_gradients)
-        latent_protos = self.backbone(protos)
-        self.backbone.requires_grad_(bb_grad)
-
-        distances = self.distance_layer(latent_x, latent_protos)
-        return distances
-
-    def predict_latent(self, x, map_protos=True):
-        """Predict `x` assuming it is already embedded in the latent space.
-
-        Only the prototypes are embedded in the latent space using the
-        backbone.
-
-        """
-        self.eval()
-        with torch.no_grad():
-            protos, plabels = self.proto_layer()
-            if map_protos:
-                protos = self.backbone(protos)
-            d = self.distance_layer(x, protos)
-            y_pred = wtac(d, plabels)
-        return y_pred
-
-
-class LVQMLN(SiameseGLVQ):
-    """Learning Vector Quantization Multi-Layer Network.
-
-    GLVQ model that applies an arbitrary transformation on the inputs, BUT NOT
-    on the prototypes before computing the distances between them. This of
-    course, means that the prototypes no longer live the input space, but
-    rather in the embedding space.
-
-    """
-
-    def compute_distances(self, x):
-        latent_protos, _ = self.proto_layer()
-        latent_x = self.backbone(x)
-        distances = self.distance_layer(latent_x, latent_protos)
-        return distances
-
-
-class GRLVQ(SiameseGLVQ):
-    """Generalized Relevance Learning Vector Quantization.
-
-    Implemented as a Siamese network with a linear transformation backbone.
-
-    TODO Make a RelevanceLayer. `bb_lr` is ignored otherwise.
-
-    """
-    _relevances: torch.Tensor
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        # Additional parameters
-        relevances = torch.ones(self.hparams["input_dim"], device=self.device)
-        self.register_parameter("_relevances", Parameter(relevances))
-
-        # Override the backbone
-        self.backbone = LambdaLayer(self._apply_relevances,
-                                    name="relevance scaling")
-
-    def _apply_relevances(self, x):
-        return x @ torch.diag(self._relevances)
-
-    @property
-    def relevance_profile(self):
-        return self._relevances.detach().cpu()
-
-    def extra_repr(self):
-        return f"(relevances): (shape: {tuple(self._relevances.shape)})"
-
-
-class SiameseGMLVQ(SiameseGLVQ):
-    """Generalized Matrix Learning Vector Quantization.
-
-    Implemented as a Siamese network with a linear transformation backbone.
-
-    """
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        # Override the backbone
-        omega_initializer = kwargs.get("omega_initializer",
-                                       EyeLinearTransformInitializer())
-        self.backbone = LinearTransform(
-            self.hparams["input_dim"],
-            self.hparams["latent_dim"],
-            initializer=omega_initializer,
-        )
-
-    @property
-    def omega_matrix(self):
-        return self.backbone.weights
-
-    @property
-    def lambda_matrix(self):
-        omega = self.backbone.weights  # (input_dim, latent_dim)
-        lam = omega @ omega.T
-        return lam.detach().cpu()
-
-
-class GMLVQ(GLVQ):
-    """Generalized Matrix Learning Vector Quantization.
-
-    Implemented as a regular GLVQ network that simply uses a different distance
-    function. This makes it easier to implement a localized variant.
-
-    """
-
-    # Parameters
-    _omega: torch.Tensor
-
-    def __init__(self, hparams, **kwargs):
-        distance_fn = kwargs.pop("distance_fn", omega_distance)
-        super().__init__(hparams, distance_fn=distance_fn, **kwargs)
-
-        # Additional parameters
-        omega_initializer = kwargs.get("omega_initializer",
-                                       EyeLinearTransformInitializer())
-        omega = omega_initializer.generate(self.hparams["input_dim"],
-                                           self.hparams["latent_dim"])
-        self.register_parameter("_omega", Parameter(omega))
-
-    @property
-    def omega_matrix(self):
-        return self._omega.detach().cpu()
-
-    @property
-    def lambda_matrix(self):
-        omega = self._omega.detach()  # (input_dim, latent_dim)
-        lam = omega @ omega.T
-        return lam.detach().cpu()
-
-    def compute_distances(self, x):
-        protos, _ = self.proto_layer()
-        distances = self.distance_layer(x, protos, self._omega)
-        return distances
-
-    def extra_repr(self):
-        return f"(omega): (shape: {tuple(self._omega.shape)})"
-
-
-class LGMLVQ(GMLVQ):
-    """Localized and Generalized Matrix Learning Vector Quantization."""
-
-    def __init__(self, hparams, **kwargs):
-        distance_fn = kwargs.pop("distance_fn", lomega_distance)
-        super().__init__(hparams, distance_fn=distance_fn, **kwargs)
-
-        # Re-register `_omega` to override the one from the super class.
-        omega = torch.randn(
-            self.num_prototypes,
-            self.hparams["input_dim"],
-            self.hparams["latent_dim"],
-            device=self.device,
-        )
-        self.register_parameter("_omega", Parameter(omega))
-
-
-class GTLVQ(LGMLVQ):
-    """Localized and Generalized Tangent Learning Vector Quantization."""
-
-    def __init__(self, hparams, **kwargs):
-        distance_fn = kwargs.pop("distance_fn", ltangent_distance)
-        super().__init__(hparams, distance_fn=distance_fn, **kwargs)
-
-        omega_initializer = kwargs.get("omega_initializer")
-
-        if omega_initializer is not None:
-            subspace = omega_initializer.generate(
-                self.hparams["input_dim"],
-                self.hparams["latent_dim"],
-            )
-            omega = torch.repeat_interleave(
-                subspace.unsqueeze(0),
-                self.num_prototypes,
-                dim=0,
-            )
-        else:
-            omega = torch.rand(
-                self.num_prototypes,
-                self.hparams["input_dim"],
-                self.hparams["latent_dim"],
-                device=self.device,
-            )
-
-        # Re-register `_omega` to override the one from the super class.
-        self.register_parameter("_omega", Parameter(omega))
-
-    def on_train_batch_end(self, outputs, batch, batch_idx):
-        with torch.no_grad():
-            self._omega.copy_(orthogonalization(self._omega))
-
-
-class SiameseGTLVQ(SiameseGLVQ, GTLVQ):
-    """Generalized Tangent Learning Vector Quantization.
-
-    Implemented as a Siamese network with a linear transformation backbone.
-
-    """
-
-
-class GLVQ1(GLVQ):
-    """Generalized Learning Vector Quantization 1."""
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-        self.loss = LossLayer(lvq1_loss)
-        self.optimizer = torch.optim.SGD
-
-
-class GLVQ21(GLVQ):
-    """Generalized Learning Vector Quantization 2.1."""
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-        self.loss = LossLayer(lvq21_loss)
-        self.optimizer = torch.optim.SGD
-
-
-class ImageGLVQ(ImagePrototypesMixin, GLVQ):
-    """GLVQ for training on image data.
-
-    GLVQ model that constrains the prototypes to the range [0, 1] by clamping
-    after updates.
-
-    """
-
-
-class ImageGMLVQ(ImagePrototypesMixin, GMLVQ):
-    """GMLVQ for training on image data.
-
-    GMLVQ model that constrains the prototypes to the range [0, 1] by clamping
-    after updates.
-
-    """
-
-
-class ImageGTLVQ(ImagePrototypesMixin, GTLVQ):
-    """GTLVQ for training on image data.
-
-    GTLVQ model that constrains the prototypes to the range [0, 1] by clamping
-    after updates.
-
-    """
-
-    def on_train_batch_end(self, outputs, batch, batch_idx):
-        """Constrain the components to the range [0, 1] by clamping after updates."""
-        self.proto_layer.components.data.clamp_(0.0, 1.0)
-        with torch.no_grad():
-            self._omega.copy_(orthogonalization(self._omega))

src/prototorch/models/knn.py

@@ -1,45 +0,0 @@
-"""ProtoTorch KNN model."""
-
-import warnings
-
-from prototorch.core.competitions import KNNC
-from prototorch.core.components import LabeledComponents
-from prototorch.core.initializers import (
-    LiteralCompInitializer,
-    LiteralLabelsInitializer,
-)
-from prototorch.utils.utils import parse_data_arg
-
-from .abstract import SupervisedPrototypeModel
-
-
-class KNN(SupervisedPrototypeModel):
-    """K-Nearest-Neighbors classification algorithm."""
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, skip_proto_layer=True, **kwargs)
-
-        # Default hparams
-        self.hparams.setdefault("k", 1)
-
-        data = kwargs.get("data", None)
-        if data is None:
-            raise ValueError("KNN requires data, but was not provided!")
-        data, targets = parse_data_arg(data)
-
-        # Layers
-        self.proto_layer = LabeledComponents(
-            distribution=len(data) * [1],
-            components_initializer=LiteralCompInitializer(data),
-            labels_initializer=LiteralLabelsInitializer(targets))
-        self.competition_layer = KNNC(k=self.hparams.k)
-
-    def training_step(self, train_batch, batch_idx):
-        return 1  # skip training step
-
-    def on_train_batch_start(self, train_batch, batch_idx):
-        warnings.warn("k-NN has no training, skipping!")
-        return -1
-
-    def configure_optimizers(self):
-        return None

src/prototorch/models/lvq.py

@@ -1,128 +0,0 @@
-"""LVQ models that are optimized using non-gradient methods."""
-
-import logging
-
-from prototorch.core.losses import _get_dp_dm
-from prototorch.nn.activations import get_activation
-from prototorch.nn.wrappers import LambdaLayer
-
-from .abstract import NonGradientMixin
-from .glvq import GLVQ
-
-
-class LVQ1(NonGradientMixin, GLVQ):
-    """Learning Vector Quantization 1."""
-
-    def training_step(self, train_batch, batch_idx):
-        protos, plables = self.proto_layer()
-        x, y = train_batch
-        dis = self.compute_distances(x)
-        # TODO Vectorized implementation
-
-        for xi, yi in zip(x, y):
-            d = self.compute_distances(xi.view(1, -1))
-            preds = self.competition_layer(d, plabels)
-            w = d.argmin(1)
-            if yi == preds:
-                shift = xi - protos[w]
-            else:
-                shift = protos[w] - xi
-            updated_protos = protos + 0.0
-            updated_protos[w] = protos[w] + (self.hparams.lr * shift)
-            self.proto_layer.load_state_dict({"_components": updated_protos},
-                                             strict=False)
-
-        logging.debug(f"dis={dis}")
-        logging.debug(f"y={y}")
-        # Logging
-        self.log_acc(dis, y, tag="train_acc")
-
-        return None
-
-
-class LVQ21(NonGradientMixin, GLVQ):
-    """Learning Vector Quantization 2.1."""
-
-    def training_step(self, train_batch, batch_idx):
-        protos, plabels = self.proto_layer()
-
-        x, y = train_batch
-        dis = self.compute_distances(x)
-        # TODO Vectorized implementation
-
-        for xi, yi in zip(x, y):
-            xi = xi.view(1, -1)
-            yi = yi.view(1, )
-            d = self.compute_distances(xi)
-            (_, wp), (_, wn) = _get_dp_dm(d, yi, plabels, with_indices=True)
-            shiftp = xi - protos[wp]
-            shiftn = protos[wn] - xi
-            updated_protos = protos + 0.0
-            updated_protos[wp] = protos[wp] + (self.hparams.lr * shiftp)
-            updated_protos[wn] = protos[wn] + (self.hparams.lr * shiftn)
-            self.proto_layer.load_state_dict({"_components": updated_protos},
-                                             strict=False)
-
-        # Logging
-        self.log_acc(dis, y, tag="train_acc")
-
-        return None
-
-
-class MedianLVQ(NonGradientMixin, GLVQ):
-    """Median LVQ
-
-    # TODO Avoid computing distances over and over
-
-    """
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        self.transfer_layer = LambdaLayer(
-            get_activation(self.hparams.transfer_fn))
-
-    def _f(self, x, y, protos, plabels):
-        d = self.distance_layer(x, protos)
-        dp, dm = _get_dp_dm(d, y, plabels)
-        mu = (dp - dm) / (dp + dm)
-        invmu = -1.0 * mu
-        f = self.transfer_layer(invmu, beta=self.hparams.transfer_beta) + 1.0
-        return f
-
-    def expectation(self, x, y, protos, plabels):
-        f = self._f(x, y, protos, plabels)
-        gamma = f / f.sum()
-        return gamma
-
-    def lower_bound(self, x, y, protos, plabels, gamma):
-        f = self._f(x, y, protos, plabels)
-        lower_bound = (gamma * f.log()).sum()
-        return lower_bound
-
-    def training_step(self, train_batch, batch_idx):
-        protos, plabels = self.proto_layer()
-
-        x, y = train_batch
-        dis = self.compute_distances(x)
-
-        for i, _ in enumerate(protos):
-            # Expectation step
-            gamma = self.expectation(x, y, protos, plabels)
-            lower_bound = self.lower_bound(x, y, protos, plabels, gamma)
-
-            # Maximization step
-            _protos = protos + 0
-            for k, xk in enumerate(x):
-                _protos[i] = xk
-                _lower_bound = self.lower_bound(x, y, _protos, plabels, gamma)
-                if _lower_bound > lower_bound:
-                    logging.debug(f"Updating prototype {i} to data {k}...")
-                    self.proto_layer.load_state_dict({"_components": _protos},
-                                                     strict=False)
-                    break
-
-        # Logging
-        self.log_acc(dis, y, tag="train_acc")
-
-        return None

src/prototorch/models/probabilistic.py

@@ -1,131 +0,0 @@
-"""Probabilistic GLVQ methods"""
-
-import torch
-from prototorch.core.losses import nllr_loss, rslvq_loss
-from prototorch.core.pooling import (
-    stratified_min_pooling,
-    stratified_sum_pooling,
-)
-from prototorch.nn.wrappers import LossLayer
-
-from .extras import GaussianPrior, RankScaledGaussianPrior
-from .glvq import GLVQ, SiameseGMLVQ
-
-
-class CELVQ(GLVQ):
-    """Cross-Entropy Learning Vector Quantization."""
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        # Loss
-        self.loss = torch.nn.CrossEntropyLoss()
-
-    def shared_step(self, batch, batch_idx):
-        x, y = batch
-        out = self.compute_distances(x)  # [None, num_protos]
-        _, plabels = self.proto_layer()
-        winning = stratified_min_pooling(out, plabels)  # [None, num_classes]
-        probs = -1.0 * winning
-        batch_loss = self.loss(probs, y.long())
-        loss = batch_loss.sum()
-        return out, loss
-
-
-class ProbabilisticLVQ(GLVQ):
-
-    def __init__(self, hparams, rejection_confidence=0.0, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        self.rejection_confidence = rejection_confidence
-        self._conditional_distribution = None
-
-    def forward(self, x):
-        distances = self.compute_distances(x)
-
-        conditional = self.conditional_distribution(distances)
-        prior = (1. / self.num_prototypes) * torch.ones(self.num_prototypes,
-                                                        device=self.device)
-        posterior = conditional * prior
-
-        plabels = self.proto_layer._labels
-        if isinstance(plabels, torch.LongTensor) or isinstance(
-                plabels, torch.cuda.LongTensor):  # type: ignore
-            y_pred = stratified_sum_pooling(posterior, plabels)  # type: ignore
-        else:
-            raise ValueError("Labels must be LongTensor.")
-
-        return y_pred
-
-    def predict(self, x):
-        y_pred = self.forward(x)
-        confidence, prediction = torch.max(y_pred, dim=1)
-        prediction[confidence < self.rejection_confidence] = -1
-        return prediction
-
-    def training_step(self, batch, batch_idx):
-        x, y = batch
-        out = self.forward(x)
-        _, plabels = self.proto_layer()
-        batch_loss = self.loss(out, y, plabels)
-        loss = batch_loss.sum()
-        return loss
-
-    def conditional_distribution(self, distances):
-        """Conditional distribution of distances."""
-        if self._conditional_distribution is None:
-            raise ValueError("Conditional distribution is not set.")
-        return self._conditional_distribution(distances)
-
-
-class SLVQ(ProbabilisticLVQ):
-    """Soft Learning Vector Quantization."""
-
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-
-        # Default hparams
-        self.hparams.setdefault("variance", 1.0)
-        variance = self.hparams.get("variance")
-
-        self._conditional_distribution = GaussianPrior(variance)
-        self.loss = LossLayer(nllr_loss)
-
-
-class RSLVQ(ProbabilisticLVQ):
-    """Robust Soft Learning Vector Quantization."""
-
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-
-        # Default hparams
-        self.hparams.setdefault("variance", 1.0)
-        variance = self.hparams.get("variance")
-
-        self._conditional_distribution = GaussianPrior(variance)
-        self.loss = LossLayer(rslvq_loss)
-
-
-class PLVQ(ProbabilisticLVQ, SiameseGMLVQ):
-    """Probabilistic Learning Vector Quantization.
-
-    TODO: Use Backbone LVQ instead
-    """
-
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-
-        # Default hparams
-        self.hparams.setdefault("lambda", 1.0)
-        lam = self.hparams.get("lambda", 1.0)
-
-        self.conditional_distribution = RankScaledGaussianPrior(lam)
-        self.loss = torch.nn.KLDivLoss()
-
-    # FIXME
-    # def training_step(self, batch, batch_idx):
-    #     x, y = batch
-    #     y_pred = self(x)
-    #     batch_loss = self.loss(y_pred, y)
-    #     loss = batch_loss.sum()
-    #     return loss

src/prototorch/models/unsupervised.py

@@ -1,154 +0,0 @@
-"""Unsupervised prototype learning algorithms."""
-
-import numpy as np
-import torch
-from prototorch.core.competitions import wtac
-from prototorch.core.distances import squared_euclidean_distance
-from prototorch.core.losses import NeuralGasEnergy
-
-from .abstract import NonGradientMixin, UnsupervisedPrototypeModel
-from .callbacks import GNGCallback
-from .extras import ConnectionTopology
-
-
-class KohonenSOM(NonGradientMixin, UnsupervisedPrototypeModel):
-    """Kohonen Self-Organizing-Map.
-
-    TODO Allow non-2D grids
-
-    """
-    _grid: torch.Tensor
-
-    def __init__(self, hparams, **kwargs):
-        h, w = hparams.get("shape")
-        # Ignore `num_prototypes`
-        hparams["num_prototypes"] = h * w
-        distance_fn = kwargs.pop("distance_fn", squared_euclidean_distance)
-        super().__init__(hparams, distance_fn=distance_fn, **kwargs)
-
-        # Hyperparameters
-        self.save_hyperparameters(hparams)
-
-        # Default hparams
-        self.hparams.setdefault("alpha", 0.3)
-        self.hparams.setdefault("sigma", max(h, w) / 2.0)
-
-        # Additional parameters
-        x, y = torch.arange(h), torch.arange(w)
-        grid = torch.stack(torch.meshgrid(x, y, indexing="ij"), dim=-1)
-        self.register_buffer("_grid", grid)
-        self._sigma = self.hparams.sigma
-        self._lr = self.hparams.lr
-
-    def predict_from_distances(self, distances):
-        grid = self._grid.view(-1, 2)
-        wp = wtac(distances, grid)
-        return wp
-
-    def training_step(self, train_batch, batch_idx):
-        # x = train_batch
-        # TODO Check if the batch has labels
-        x = train_batch[0]
-        d = self.compute_distances(x)
-        wp = self.predict_from_distances(d)
-        grid = self._grid.view(-1, 2)
-        gd = squared_euclidean_distance(wp, grid)
-        nh = torch.exp(-gd / self._sigma**2)
-        protos = self.proto_layer()
-        diff = x.unsqueeze(dim=1) - protos
-        delta = self._lr * self.hparams.alpha * nh.unsqueeze(-1) * diff
-        updated_protos = protos + delta.sum(dim=0)
-        self.proto_layer.load_state_dict(
-            {"_components": updated_protos},
-            strict=False,
-        )
-
-    def on_training_epoch_end(self, training_step_outputs):
-        self._sigma = self.hparams.sigma * np.exp(
-            -self.current_epoch / self.trainer.max_epochs)
-
-    def extra_repr(self):
-        return f"(grid): (shape: {tuple(self._grid.shape)})"
-
-
-class HeskesSOM(UnsupervisedPrototypeModel):
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-    def training_step(self, train_batch, batch_idx):
-        # TODO Implement me!
-        raise NotImplementedError()
-
-
-class NeuralGas(UnsupervisedPrototypeModel):
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        # Hyperparameters
-        self.save_hyperparameters(hparams)
-
-        # Default hparams
-        self.hparams.setdefault("age_limit", 10)
-        self.hparams.setdefault("lm", 1)
-
-        self.energy_layer = NeuralGasEnergy(lm=self.hparams["lm"])
-        self.topology_layer = ConnectionTopology(
-            agelimit=self.hparams["age_limit"],
-            num_prototypes=self.hparams["num_prototypes"],
-        )
-
-    def training_step(self, train_batch, batch_idx):
-        # x = train_batch
-        # TODO Check if the batch has labels
-        x = train_batch[0]
-        d = self.compute_distances(x)
-        loss, _ = self.energy_layer(d)
-        self.topology_layer(d)
-        self.log("loss", loss)
-        return loss
-
-
-class GrowingNeuralGas(NeuralGas):
-    errors: torch.Tensor
-
-    def __init__(self, hparams, **kwargs):
-        super().__init__(hparams, **kwargs)
-
-        # Defaults
-        self.hparams.setdefault("step_reduction", 0.5)
-        self.hparams.setdefault("insert_reduction", 0.1)
-        self.hparams.setdefault("insert_freq", 10)
-
-        errors = torch.zeros(
-            self.hparams["num_prototypes"],
-            device=self.device,
-        )
-        self.register_buffer("errors", errors)
-
-    def training_step(self, train_batch, _batch_idx):
-        # x = train_batch
-        # TODO Check if the batch has labels
-        x = train_batch[0]
-        d = self.compute_distances(x)
-        loss, order = self.energy_layer(d)
-        winner = order[:, 0]
-        mask = torch.zeros_like(d)
-        mask[torch.arange(len(mask)), winner] = 1.0
-        dp = d * mask
-
-        self.errors += torch.sum(dp * dp)
-        self.errors *= self.hparams["step_reduction"]
-
-        self.topology_layer(d)
-        self.log("loss", loss)
-        return loss
-
-    def configure_callbacks(self):
-        return [
-            GNGCallback(
-                reduction=self.hparams["insert_reduction"],
-                freq=self.hparams["insert_freq"],
-            )
-        ]

tests/test_models.py

@@ -1,193 +1,13 @@
 """prototorch.models test suite."""
 
-import prototorch.models
+import prototorch as pt
+from prototorch.models.library import GLVQ
 
 
 def test_glvq_model_build():
-    model = prototorch.models.GLVQ(
+    hparams = GLVQ.HyperParameters(
-        {"distribution": (3, 2)},
+        distribution=dict(num_classes=2, per_class=1),
-        prototypes_initializer=prototorch.initializers.RNCI(2),
+        component_initializer=pt.initializers.RNCI(2),
     )
 
-
+    model = GLVQ(hparams=hparams)
-def test_glvq1_model_build():
-    model = prototorch.models.GLVQ1(
-        {"distribution": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_glvq21_model_build():
-    model = prototorch.models.GLVQ1(
-        {"distribution": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_gmlvq_model_build():
-    model = prototorch.models.GMLVQ(
-        {
-            "distribution": (3, 2),
-            "input_dim": 2,
-            "latent_dim": 2,
-        },
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_grlvq_model_build():
-    model = prototorch.models.GRLVQ(
-        {
-            "distribution": (3, 2),
-            "input_dim": 2,
-        },
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_gtlvq_model_build():
-    model = prototorch.models.GTLVQ(
-        {
-            "distribution": (3, 2),
-            "input_dim": 4,
-            "latent_dim": 2,
-        },
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_lgmlvq_model_build():
-    model = prototorch.models.LGMLVQ(
-        {
-            "distribution": (3, 2),
-            "input_dim": 4,
-            "latent_dim": 2,
-        },
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_image_glvq_model_build():
-    model = prototorch.models.ImageGLVQ(
-        {"distribution": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(16),
-    )
-
-
-def test_image_gmlvq_model_build():
-    model = prototorch.models.ImageGMLVQ(
-        {
-            "distribution": (3, 2),
-            "input_dim": 16,
-            "latent_dim": 2,
-        },
-        prototypes_initializer=prototorch.initializers.RNCI(16),
-    )
-
-
-def test_image_gtlvq_model_build():
-    model = prototorch.models.ImageGMLVQ(
-        {
-            "distribution": (3, 2),
-            "input_dim": 16,
-            "latent_dim": 2,
-        },
-        prototypes_initializer=prototorch.initializers.RNCI(16),
-    )
-
-
-def test_siamese_glvq_model_build():
-    model = prototorch.models.SiameseGLVQ(
-        {"distribution": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(4),
-    )
-
-
-def test_siamese_gmlvq_model_build():
-    model = prototorch.models.SiameseGMLVQ(
-        {
-            "distribution": (3, 2),
-            "input_dim": 4,
-            "latent_dim": 2,
-        },
-        prototypes_initializer=prototorch.initializers.RNCI(4),
-    )
-
-
-def test_siamese_gtlvq_model_build():
-    model = prototorch.models.SiameseGTLVQ(
-        {
-            "distribution": (3, 2),
-            "input_dim": 4,
-            "latent_dim": 2,
-        },
-        prototypes_initializer=prototorch.initializers.RNCI(4),
-    )
-
-
-def test_knn_model_build():
-    train_ds = prototorch.datasets.Iris(dims=[0, 2])
-    model = prototorch.models.KNN(dict(k=3), data=train_ds)
-
-
-def test_lvq1_model_build():
-    model = prototorch.models.LVQ1(
-        {"distribution": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_lvq21_model_build():
-    model = prototorch.models.LVQ21(
-        {"distribution": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_median_lvq_model_build():
-    model = prototorch.models.MedianLVQ(
-        {"distribution": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_celvq_model_build():
-    model = prototorch.models.CELVQ(
-        {"distribution": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_rslvq_model_build():
-    model = prototorch.models.RSLVQ(
-        {"distribution": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_slvq_model_build():
-    model = prototorch.models.SLVQ(
-        {"distribution": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_growing_neural_gas_model_build():
-    model = prototorch.models.GrowingNeuralGas(
-        {"num_prototypes": 5},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_kohonen_som_model_build():
-    model = prototorch.models.KohonenSOM(
-        {"shape": (3, 2)},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )
-
-
-def test_neural_gas_model_build():
-    model = prototorch.models.NeuralGas(
-        {"num_prototypes": 5},
-        prototypes_initializer=prototorch.initializers.RNCI(2),
-    )