oodeel 0.1.1__py3-none-any.whl → 0.3.0__py3-none-any.whl

oodeel/utils/tf_training_tools.py

@@ -35,6 +35,27 @@ from ..types import Optional
 from ..types import Union
 
 
+def get_toy_mlp(input_shape: tuple, num_classes: int) -> tf.keras.Model:
+    """Basic keras MLP classifier for toy datasets.
+
+    Args:
+        input_shape (tuple): Input data shape.
+        num_classes (int): Number of classes for the classification task.
+
+    Returns:
+        tf.keras.Model: model
+    """
+    return tf.keras.models.Sequential(
+        [
+            tf.keras.layers.Input(shape=input_shape),
+            tf.keras.layers.Flatten(),
+            tf.keras.layers.Dense(300, activation="relu"),
+            tf.keras.layers.Dense(150, activation="relu"),
+            tf.keras.layers.Dense(num_classes, activation="softmax"),
+        ]
+    )
+
+
 def get_toy_keras_convnet(num_classes: int) -> tf.keras.Model:
     """Basic keras convolutional classifier for toy datasets.
 
@@ -60,8 +81,8 @@ def get_toy_keras_convnet(num_classes: int) -> tf.keras.Model:
 def train_tf_model(
     train_data: tf.data.Dataset,
     model: Union[tf.keras.Model, str],
-    input_shape: tuple,
-    num_classes: int,
+    input_shape: tuple = None,
+    num_classes: int = None,
     batch_size: int = 128,
     epochs: int = 50,
     loss: str = "sparse_categorical_crossentropy",
@@ -80,7 +101,7 @@ def train_tf_model(
     Args:
         train_data (tf.data.Dataset): training dataset.
         model (Union[tf.keras.Model, str]): if a string is provided, must be a model
-            from tf.keras.applications or "toy_convnet"
+            from tf.keras.applications or "toy_convnet" or "toy_mlp"
         input_shape (tuple): Shape of the input images.
         num_classes (int): Number of output classes.
         batch_size (int, optional): Defaults to 128.
@@ -120,6 +141,8 @@ def train_tf_model(
     elif isinstance(model, str):
         if model == "toy_convnet":
             model = get_toy_keras_convnet(num_classes)
+        elif model == "toy_mlp":
+            model = get_toy_mlp(input_shape, num_classes)
         else:
             weights = "imagenet" if imagenet_pretrained else None
             backbone = getattr(tf.keras.applications, model)(

oodeel/utils/torch_operator.py

@@ -69,12 +69,24 @@ class TorchOperator(Operator):
         return torch.argmax(tensor, dim=dim)
 
     @staticmethod
-    def max(tensor: TensorType, dim: Optional[int] = None) -> torch.Tensor:
+    def max(
+        tensor: TensorType, dim: Optional[int] = None, keepdim: Optional[bool] = False
+    ) -> torch.Tensor:
         """Max function"""
         if dim is None:
             return torch.max(tensor)
         else:
-            return torch.max(tensor, dim)[0]
+            return torch.max(tensor, dim, keepdim=keepdim)[0]
+
+    @staticmethod
+    def min(
+        tensor: TensorType, dim: Optional[int] = None, keepdim: bool = False
+    ) -> torch.Tensor:
+        """Min function"""
+        if dim is None:
+            return torch.min(tensor)
+        else:
+            return torch.min(tensor, dim, keepdim=keepdim)[0]
 
     @staticmethod
     def one_hot(tensor: TensorType, num_classes: int) -> torch.Tensor:
@@ -167,13 +179,18 @@ class TorchOperator(Operator):
     def from_numpy(self, arr: np.ndarray) -> torch.Tensor:
         "Convert a NumPy array to a tensor"
         # TODO change dtype
-        return torch.from_numpy(arr).double().to(self._device)
+        return torch.tensor(arr).to(self._device)
 
     @staticmethod
-    def transpose(tensor: TensorType) -> torch.Tensor:
+    def t(tensor: TensorType) -> torch.Tensor:
         "Transpose function for tensor of rank 2"
         return tensor.t()
 
+    @staticmethod
+    def permute(tensor: TensorType, dims) -> torch.Tensor:
+        "Transpose function for tensor of rank 2"
+        return torch.permute(tensor, dims)
+
     @staticmethod
     def diag(tensor: TensorType) -> torch.Tensor:
         "Diagonal function: return the diagonal of a 2D tensor"
@@ -215,3 +232,57 @@ class TorchOperator(Operator):
     def relu(tensor: TensorType) -> torch.Tensor:
         "Apply relu to a tensor"
         return torch.nn.functional.relu(tensor)
+
+    @staticmethod
+    def einsum(equation: str, *tensors: TensorType) -> torch.Tensor:
+        "Computes the einsum between tensors following equation"
+        return torch.einsum(equation, *tensors)
+
+    @staticmethod
+    def tril(tensor: TensorType, diagonal: int = 0) -> torch.Tensor:
+        "Set the upper triangle of the matrix formed by the last two dimensions of"
+        "tensor to zero"
+        return torch.tril(tensor, diagonal)
+
+    @staticmethod
+    def sum(tensor: TensorType, dim: Union[tuple, list, int] = None) -> torch.Tensor:
+        "sum along dim"
+        return torch.sum(tensor, dim)
+
+    @staticmethod
+    def unsqueeze(tensor: TensorType, dim: int) -> torch.Tensor:
+        "unsqueeze along dim"
+        return torch.unsqueeze(tensor, dim)
+
+    @staticmethod
+    def squeeze(tensor: TensorType, dim: int = None) -> torch.Tensor:
+        "squeeze along dim"
+
+        if dim is None:
+            return torch.squeeze(tensor)
+
+        return torch.squeeze(tensor, dim)
+
+    @staticmethod
+    def abs(tensor: TensorType) -> torch.Tensor:
+        "compute absolute value"
+        return torch.abs(tensor)
+
+    @staticmethod
+    def where(
+        condition: TensorType,
+        input: Union[TensorType, float],
+        other: Union[TensorType, float],
+    ) -> torch.Tensor:
+        "Applies where function , to condition"
+        return torch.where(condition, input, other)
+
+    @staticmethod
+    def avg_pool_2d(tensor: TensorType) -> torch.Tensor:
+        """Perform avg pool in 2d as in torch.nn.functional.adaptive_avg_pool2d"""
+        return torch.mean(tensor, dim=(-2, -1))
+
+    @staticmethod
+    def log(tensor: TensorType) -> torch.Tensor:
+        """Perform log"""
+        return torch.log(tensor)

oodeel/utils/torch_training_tools.py

@@ -35,6 +35,31 @@ from ..types import Optional
 from ..types import Union
 
 
+class ToyTorchMLP(nn.Sequential):
+    """Basic torch MLP classifier for toy datasets.
+
+    Args:
+        input_shape (tuple): Input data shape.
+        num_classes (int): Number of classes for the classification task.
+    """
+
+    def __init__(self, input_shape: tuple, num_classes: int):
+        self.input_shape = input_shape
+
+        # build toy mlp
+        mlp_modules = OrderedDict(
+            [
+                ("flatten", nn.Flatten()),
+                ("dense1", nn.Linear(np.prod(input_shape), 300)),
+                ("relu1", nn.ReLU()),
+                ("dense2", nn.Linear(300, 150)),
+                ("relu2", nn.ReLU()),
+                ("fc1", nn.Linear(150, num_classes)),
+            ]
+        )
+        super().__init__(mlp_modules)
+
+
 class ToyTorchConvnet(nn.Sequential):
     """Basic torch convolutional classifier for toy datasets.
 
@@ -106,7 +131,7 @@ def train_torch_model(
     Args:
         train_data (DataLoader): train dataloader
         model (Union[nn.Module, str]): if a string is provided, must be a model from
-            torchvision.models or "toy_convnet".
+            torchvision.models or "toy_convnet" or "toy_mlp.
         num_classes (int): Number of output classes.
         epochs (int, optional): Defaults to 50.
         loss (str, optional): Defaults to "CrossEntropyLoss".
@@ -136,8 +161,12 @@ def train_torch_model(
     elif isinstance(model, str):
         if model == "toy_convnet":
             # toy model
-            input_shape = next(iter(train_data))[0].shape[1:]
+            input_shape = tuple(next(iter(train_data))[0].shape[1:])
             model = ToyTorchConvnet(input_shape, num_classes).to(device)
+        elif model == "toy_mlp":
+            # toy model
+            input_shape = tuple(next(iter(train_data))[0].shape[1:])
+            model = ToyTorchMLP(input_shape, num_classes).to(device)
         else:
             # torchvision model
             model = getattr(torchvision.models, model)(

{oodeel-0.1.1.dist-info → oodeel-0.3.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: oodeel
-Version: 0.1.1
+Version: 0.3.0
 Summary: Simple, compact, and hackable post-hoc deep OOD detection for alreadytrained tensorflow or pytorch image classifiers.
 Author: DEEL Core Team
 Author-email: paul.novello@irt-saintexupery.com
@@ -12,104 +12,126 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: faiss-cpu
+Requires-Dist: faiss_cpu
 Requires-Dist: numpy
-Requires-Dist: scikit-learn
+Requires-Dist: scikit_learn
 Requires-Dist: scipy
 Requires-Dist: setuptools
 Requires-Dist: matplotlib
 Requires-Dist: pandas
 Requires-Dist: seaborn
 Requires-Dist: plotly
+Requires-Dist: tqdm
 Provides-Extra: dev
-Requires-Dist: mypy ; extra == 'dev'
-Requires-Dist: ipywidgets ; extra == 'dev'
-Requires-Dist: mkdocs-jupyter ; extra == 'dev'
-Requires-Dist: mkdocstrings-python ; extra == 'dev'
-Requires-Dist: flake8 ; extra == 'dev'
-Requires-Dist: setuptools ; extra == 'dev'
-Requires-Dist: pre-commit ; extra == 'dev'
-Requires-Dist: tox ; extra == 'dev'
-Requires-Dist: black ; extra == 'dev'
-Requires-Dist: ipython ; extra == 'dev'
-Requires-Dist: ipykernel ; extra == 'dev'
-Requires-Dist: pytest ; extra == 'dev'
-Requires-Dist: pylint ; extra == 'dev'
-Requires-Dist: mkdocs ; extra == 'dev'
-Requires-Dist: mkdocs-material ; extra == 'dev'
-Requires-Dist: mkdocstrings ; extra == 'dev'
-Requires-Dist: mknotebooks ; extra == 'dev'
-Requires-Dist: bump2version ; extra == 'dev'
-Requires-Dist: docsig ; extra == 'dev'
-Requires-Dist: no-implicit-optional ; extra == 'dev'
-Requires-Dist: tensorflow ; extra == 'dev'
-Requires-Dist: tensorflow-datasets ; extra == 'dev'
-Requires-Dist: tensorflow-probability ; extra == 'dev'
-Requires-Dist: timm ; extra == 'dev'
-Requires-Dist: torch ; extra == 'dev'
-Requires-Dist: torchvision ; extra == 'dev'
-Provides-Extra: docs
-Requires-Dist: mkdocs ; extra == 'docs'
-Requires-Dist: mkdocs-material ; extra == 'docs'
-Requires-Dist: mkdocstrings ; extra == 'docs'
-Requires-Dist: mknotebooks ; extra == 'docs'
-Requires-Dist: ipython ; extra == 'docs'
-Provides-Extra: tensorflow
-Requires-Dist: tensorflow ; extra == 'tensorflow'
-Requires-Dist: tensorflow-datasets ; extra == 'tensorflow'
-Requires-Dist: tensorflow-probability ; extra == 'tensorflow'
+Requires-Dist: mypy; extra == "dev"
+Requires-Dist: ipywidgets; extra == "dev"
+Requires-Dist: mkdocs-jupyter; extra == "dev"
+Requires-Dist: mkdocstrings-python; extra == "dev"
+Requires-Dist: flake8; extra == "dev"
+Requires-Dist: setuptools; extra == "dev"
+Requires-Dist: pre-commit; extra == "dev"
+Requires-Dist: tox; extra == "dev"
+Requires-Dist: black; extra == "dev"
+Requires-Dist: ruff; extra == "dev"
+Requires-Dist: ipython; extra == "dev"
+Requires-Dist: ipykernel; extra == "dev"
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: pylint; extra == "dev"
+Requires-Dist: mypy; extra == "dev"
+Requires-Dist: mkdocs; extra == "dev"
+Requires-Dist: mkdocs-material; extra == "dev"
+Requires-Dist: mkdocstrings; extra == "dev"
+Requires-Dist: mknotebooks; extra == "dev"
+Requires-Dist: mike; extra == "dev"
+Requires-Dist: bump2version; extra == "dev"
+Requires-Dist: docsig; extra == "dev"
+Requires-Dist: no_implicit_optional; extra == "dev"
+Requires-Dist: numpy==1.26.4; extra == "dev"
+Requires-Dist: tensorflow==2.11.0; extra == "dev"
+Requires-Dist: tensorflow_datasets; extra == "dev"
+Requires-Dist: tensorflow_probability==0.19.0; extra == "dev"
+Requires-Dist: timm; extra == "dev"
+Requires-Dist: torch==1.13.1; extra == "dev"
+Requires-Dist: torchvision==0.14.1; extra == "dev"
 Provides-Extra: tensorflow-dev
-Requires-Dist: mypy ; extra == 'tensorflow-dev'
-Requires-Dist: ipywidgets ; extra == 'tensorflow-dev'
-Requires-Dist: mkdocs-jupyter ; extra == 'tensorflow-dev'
-Requires-Dist: mkdocstrings-python ; extra == 'tensorflow-dev'
-Requires-Dist: flake8 ; extra == 'tensorflow-dev'
-Requires-Dist: setuptools ; extra == 'tensorflow-dev'
-Requires-Dist: pre-commit ; extra == 'tensorflow-dev'
-Requires-Dist: tox ; extra == 'tensorflow-dev'
-Requires-Dist: black ; extra == 'tensorflow-dev'
-Requires-Dist: ipython ; extra == 'tensorflow-dev'
-Requires-Dist: ipykernel ; extra == 'tensorflow-dev'
-Requires-Dist: pytest ; extra == 'tensorflow-dev'
-Requires-Dist: pylint ; extra == 'tensorflow-dev'
-Requires-Dist: mkdocs ; extra == 'tensorflow-dev'
-Requires-Dist: mkdocs-material ; extra == 'tensorflow-dev'
-Requires-Dist: mkdocstrings ; extra == 'tensorflow-dev'
-Requires-Dist: mknotebooks ; extra == 'tensorflow-dev'
-Requires-Dist: bump2version ; extra == 'tensorflow-dev'
-Requires-Dist: docsig ; extra == 'tensorflow-dev'
-Requires-Dist: no-implicit-optional ; extra == 'tensorflow-dev'
-Requires-Dist: tensorflow ; extra == 'tensorflow-dev'
-Requires-Dist: tensorflow-datasets ; extra == 'tensorflow-dev'
-Requires-Dist: tensorflow-probability ; extra == 'tensorflow-dev'
-Provides-Extra: torch
-Requires-Dist: timm ; extra == 'torch'
-Requires-Dist: torch ; extra == 'torch'
-Requires-Dist: torchvision ; extra == 'torch'
+Requires-Dist: mypy; extra == "tensorflow-dev"
+Requires-Dist: ipywidgets; extra == "tensorflow-dev"
+Requires-Dist: mkdocs-jupyter; extra == "tensorflow-dev"
+Requires-Dist: mkdocstrings-python; extra == "tensorflow-dev"
+Requires-Dist: flake8; extra == "tensorflow-dev"
+Requires-Dist: setuptools; extra == "tensorflow-dev"
+Requires-Dist: pre-commit; extra == "tensorflow-dev"
+Requires-Dist: tox; extra == "tensorflow-dev"
+Requires-Dist: black; extra == "tensorflow-dev"
+Requires-Dist: ruff; extra == "tensorflow-dev"
+Requires-Dist: ipython; extra == "tensorflow-dev"
+Requires-Dist: ipykernel; extra == "tensorflow-dev"
+Requires-Dist: pytest; extra == "tensorflow-dev"
+Requires-Dist: pylint; extra == "tensorflow-dev"
+Requires-Dist: mypy; extra == "tensorflow-dev"
+Requires-Dist: mkdocs; extra == "tensorflow-dev"
+Requires-Dist: mkdocs-material; extra == "tensorflow-dev"
+Requires-Dist: mkdocstrings; extra == "tensorflow-dev"
+Requires-Dist: mknotebooks; extra == "tensorflow-dev"
+Requires-Dist: mike; extra == "tensorflow-dev"
+Requires-Dist: bump2version; extra == "tensorflow-dev"
+Requires-Dist: docsig; extra == "tensorflow-dev"
+Requires-Dist: no_implicit_optional; extra == "tensorflow-dev"
+Requires-Dist: numpy==1.26.4; extra == "tensorflow-dev"
+Requires-Dist: tensorflow==2.11.0; extra == "tensorflow-dev"
+Requires-Dist: tensorflow_datasets; extra == "tensorflow-dev"
+Requires-Dist: tensorflow_probability==0.19.0; extra == "tensorflow-dev"
 Provides-Extra: torch-dev
-Requires-Dist: mypy ; extra == 'torch-dev'
-Requires-Dist: ipywidgets ; extra == 'torch-dev'
-Requires-Dist: mkdocs-jupyter ; extra == 'torch-dev'
-Requires-Dist: mkdocstrings-python ; extra == 'torch-dev'
-Requires-Dist: flake8 ; extra == 'torch-dev'
-Requires-Dist: setuptools ; extra == 'torch-dev'
-Requires-Dist: pre-commit ; extra == 'torch-dev'
-Requires-Dist: tox ; extra == 'torch-dev'
-Requires-Dist: black ; extra == 'torch-dev'
-Requires-Dist: ipython ; extra == 'torch-dev'
-Requires-Dist: ipykernel ; extra == 'torch-dev'
-Requires-Dist: pytest ; extra == 'torch-dev'
-Requires-Dist: pylint ; extra == 'torch-dev'
-Requires-Dist: mkdocs ; extra == 'torch-dev'
-Requires-Dist: mkdocs-material ; extra == 'torch-dev'
-Requires-Dist: mkdocstrings ; extra == 'torch-dev'
-Requires-Dist: mknotebooks ; extra == 'torch-dev'
-Requires-Dist: bump2version ; extra == 'torch-dev'
-Requires-Dist: docsig ; extra == 'torch-dev'
-Requires-Dist: no-implicit-optional ; extra == 'torch-dev'
-Requires-Dist: timm ; extra == 'torch-dev'
-Requires-Dist: torch ; extra == 'torch-dev'
-Requires-Dist: torchvision ; extra == 'torch-dev'
+Requires-Dist: mypy; extra == "torch-dev"
+Requires-Dist: ipywidgets; extra == "torch-dev"
+Requires-Dist: mkdocs-jupyter; extra == "torch-dev"
+Requires-Dist: mkdocstrings-python; extra == "torch-dev"
+Requires-Dist: flake8; extra == "torch-dev"
+Requires-Dist: setuptools; extra == "torch-dev"
+Requires-Dist: pre-commit; extra == "torch-dev"
+Requires-Dist: tox; extra == "torch-dev"
+Requires-Dist: black; extra == "torch-dev"
+Requires-Dist: ruff; extra == "torch-dev"
+Requires-Dist: ipython; extra == "torch-dev"
+Requires-Dist: ipykernel; extra == "torch-dev"
+Requires-Dist: pytest; extra == "torch-dev"
+Requires-Dist: pylint; extra == "torch-dev"
+Requires-Dist: mypy; extra == "torch-dev"
+Requires-Dist: mkdocs; extra == "torch-dev"
+Requires-Dist: mkdocs-material; extra == "torch-dev"
+Requires-Dist: mkdocstrings; extra == "torch-dev"
+Requires-Dist: mknotebooks; extra == "torch-dev"
+Requires-Dist: mike; extra == "torch-dev"
+Requires-Dist: bump2version; extra == "torch-dev"
+Requires-Dist: docsig; extra == "torch-dev"
+Requires-Dist: no_implicit_optional; extra == "torch-dev"
+Requires-Dist: numpy==1.26.4; extra == "torch-dev"
+Requires-Dist: timm; extra == "torch-dev"
+Requires-Dist: torch==1.13.1; extra == "torch-dev"
+Requires-Dist: torchvision==0.14.1; extra == "torch-dev"
+Provides-Extra: tensorflow
+Requires-Dist: tensorflow==2.11.0; extra == "tensorflow"
+Requires-Dist: tensorflow_datasets; extra == "tensorflow"
+Requires-Dist: tensorflow_probability==0.19.0; extra == "tensorflow"
+Provides-Extra: torch
+Requires-Dist: timm; extra == "torch"
+Requires-Dist: torch==1.13.1; extra == "torch"
+Requires-Dist: torchvision==0.14.1; extra == "torch"
+Provides-Extra: docs
+Requires-Dist: mkdocs; extra == "docs"
+Requires-Dist: mkdocs-material; extra == "docs"
+Requires-Dist: mkdocstrings; extra == "docs"
+Requires-Dist: mknotebooks; extra == "docs"
+Requires-Dist: ipython; extra == "docs"
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: license-file
+Dynamic: provides-extra
+Dynamic: requires-dist
+Dynamic: summary
 
 
 <!-- Banner section -->
@@ -125,29 +147,23 @@ Requires-Dist: torchvision ; extra == 'torch-dev'
 <!-- Badge section -->
 <div align="center">
     <a href="#">
-        <img src="https://img.shields.io/badge/python-3.8%2B-blue">
-    </a>
+        <img src="https://img.shields.io/badge/python-3.8%2B-blue"></a>
     <a href="https://github.com/deel-ai/oodeel/actions/workflows/python-linters.yml">
-        <img alt="Flake8" src="https://github.com/deel-ai/oodeel/actions/workflows/python-linters.yml/badge.svg">
-    </a>
+        <img alt="Flake8" src="https://github.com/deel-ai/oodeel/actions/workflows/python-linters.yml/badge.svg"></a>
     <a href="https://github.com/deel-ai/oodeel/actions/workflows/python-tests-tf.yml">
-        <img alt="Tests tf" src="https://github.com/deel-ai/oodeel/actions/workflows/python-tests-tf.yml/badge.svg">
-    </a>
+        <img alt="Tests tf" src="https://github.com/deel-ai/oodeel/actions/workflows/python-tests-tf.yml/badge.svg"></a>
     <a href="https://github.com/deel-ai/oodeel/actions/workflows/python-tests-torch.yml">
-        <img alt="Tests torch" src="https://github.com/deel-ai/oodeel/actions/workflows/python-tests-torch.yml/badge.svg">
-    </a>
+        <img alt="Tests torch" src="https://github.com/deel-ai/oodeel/actions/workflows/python-tests-torch.yml/badge.svg"></a>
     <a href="https://github.com/deel-ai/oodeel/actions/workflows/python-coverage-shield.yml">
-        <img alt="Coverage" src="https://github.com/deel-ai/oodeel/raw/gh-shields/coverage.svg">
-    </a>
+        <img alt="Coverage" src="https://github.com/deel-ai/oodeel/raw/gh-shields/coverage.svg"></a>
     <a href="https://github.com/deel-ai/oodeel/blob/master/LICENSE">
-        <img alt="License MIT" src="https://img.shields.io/badge/License-MIT-efefef">
-    </a>
+        <img alt="License MIT" src="https://img.shields.io/badge/License-MIT-efefef"></a>
 </div>
 <br>
 
 <!-- Short description of your library -->
 
-<b>Oodeel</b> is a library that performs post-hoc deep OOD detection on already trained neural network image classifiers. The philosophy of the library is to favor quality over quantity and to foster easy adoption. As a result, we provide a simple, compact and easily customizable API and carefully integrate and test each proposed baseline into a coherent framework that is designed to enable their use in tensorflow **and** pytorch. You can find the documentation [here](https://deel-ai.github.io/oodeel/).
+<b>Oodeel</b> is a library that performs post-hoc deep OOD (Out-of-Distribution) detection on already trained neural network image classifiers. The philosophy of the library is to favor quality over quantity and to foster easy adoption. As a result, we provide a simple, compact and easily customizable API and carefully integrate and test each proposed baseline into a coherent framework that is designed to enable their use in tensorflow **and** pytorch. You can find the documentation [here](https://deel-ai.github.io/oodeel/).
 
 ```python
 from oodeel.methods import MLS
@@ -167,7 +183,8 @@ scores, info = mls.score(ds) # ds is a tf.data.Dataset or a torch.DataLoader
 - [Contributing](#contributing)
 - [See Also](#see-also)
 - [Acknowledgments](#acknowledgments)
-- [Creator](#creator)
+- [Creators](#creators)
+- [Citation](#citation)
 - [License](#license)
 
 # Installation
@@ -286,15 +303,19 @@ Currently, **oodeel** includes the following baselines:
 | MSP | [A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks](http://arxiv.org/abs/1610.02136) | ICLR 2017 | avail [tensorflow & torch](docs/pages/getting_started.ipynb)|
 | Mahalanobis | [A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks](http://arxiv.org/abs/1807.03888) | NeurIPS 2018 | avail [tensorflow](docs/notebooks/tensorflow/demo_mahalanobis_tf.ipynb) or  [torch](docs/notebooks/torch/demo_mahalanobis_torch.ipynb)|
 | Energy | [Energy-based Out-of-distribution Detection](http://arxiv.org/abs/2010.03759) | NeurIPS 2020 |avail [tensorflow](docs/notebooks/tensorflow/demo_energy_tf.ipynb) or  [torch](docs/notebooks/torch/demo_energy_torch.ipynb)  |
-| Odin | [Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks](http://arxiv.org/abs/1706.02690) | ICLR 2018 |avail [tensorflow](docs/notebooks/tensorflow/demo_odin_tf.ipynb) or  [torch](docs/notebooks/torch/demo_odin_torch.ipynb) |
+| Odin | [Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks](http://arxiv.org/abs/1706.02690) | ICLR 2018 | avail [tensorflow](docs/notebooks/tensorflow/demo_odin_tf.ipynb) or  [torch](docs/notebooks/torch/demo_odin_torch.ipynb) |
 | DKNN | [Out-of-Distribution Detection with Deep Nearest Neighbors](http://arxiv.org/abs/2204.06507) | ICML 2022 | avail [tensorflow](docs/notebooks/tensorflow/demo_dknn_tf.ipynb) or  [torch](docs/notebooks/torch/demo_dknn_torch.ipynb) |
 | VIM | [ViM: Out-Of-Distribution with Virtual-logit Matching](http://arxiv.org/abs/2203.10807) | CVPR 2022 |avail [tensorflow](docs/notebooks/tensorflow/demo_vim_tf.ipynb) or  [torch](docs/notebooks/torch/demo_vim_torch.ipynb)  |
 | Entropy | [Likelihood Ratios for Out-of-Distribution Detection](https://proceedings.neurips.cc/paper/2019/hash/1e79596878b2320cac26dd792a6c51c9-Abstract.html) | NeurIPS 2019 |avail [tensorflow](docs/notebooks/tensorflow/demo_entropy_tf.ipynb) or  [torch](docs/notebooks/torch/demo_entropy_torch.ipynb)  |
 | GODIN | [Generalized ODIN: Detecting Out-of-Distribution Image Without Learning From Out-of-Distribution Data](https://ieeexplore.ieee.org/document/9156473/) | CVPR 2020 | planned |
 | ReAct | [ReAct: Out-of-distribution Detection With Rectified Activations](http://arxiv.org/abs/2111.12797) | NeurIPS 2021 | avail [tensorflow](docs/notebooks/tensorflow/demo_react_tf.ipynb) or  [torch](docs/notebooks/torch/demo_react_torch.ipynb) |
 | NMD | [Neural Mean Discrepancy for Efficient Out-of-Distribution Detection](https://openaccess.thecvf.com/content/CVPR2022/html/Dong_Neural_Mean_Discrepancy_for_Efficient_Out-of-Distribution_Detection_CVPR_2022_paper.html) | CVPR 2022 | planned |
-| Gram | [Detecting Out-of-Distribution Examples with Gram Matrices](https://proceedings.mlr.press/v119/sastry20a.html) | ICML 2020 | planned |
-
+| Gram | [Detecting Out-of-Distribution Examples with Gram Matrices](https://proceedings.mlr.press/v119/sastry20a.html) | ICML 2020 | avail [tensorflow](docs/notebooks/tensorflow/demo_gram_tf.ipynb) or  [torch](docs/notebooks/torch/demo_gram_torch.ipynb) |
+| GEN | [GEN: Pushing the Limits of Softmax-Based Out-of-Distribution Detection](https://openaccess.thecvf.com/content/CVPR2023/html/Liu_GEN_Pushing_the_Limits_of_Softmax-Based_Out-of-Distribution_Detection_CVPR_2023_paper.html) | CVPR 2023 | avail [tensorflow](docs/notebooks/tensorflow/demo_gen_tf.ipynb) or [torch](docs/notebooks/torch/demo_gen_torch.ipynb) |
+| RMDS | [A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detection](https://arxiv.org/abs/2106.09022) | preprint | avail [tensorflow](docs/notebooks/tensorflow/demo_rmds_tf.ipynb) or [torch](docs/notebooks/torch/demo_rmds_torch.ipynb) |
+| SHE | [Out-of-Distribution Detection based on In-Distribution Data Patterns Memorization with Modern Hopfield Energy](https://openreview.net/forum?id=KkazG4lgKL) | ICLR 2023 | avail [tensorflow](docs/notebooks/tensorflow/demo_she_tf.ipynb) or [torch](docs/notebooks/torch/demo_she_torch.ipynb) |
+| ASH | [Extremely Simple Activation Shaping for Out-of-Distribution Detection](http://arxiv.org/abs/2310.00227) | ICLR 2023 | avail [tensorflow](docs/notebooks/tensorflow/demo_ash_tf.ipynb) or [torch](docs/notebooks/torch/demo_ash_torch.ipynb) |
+| SCALE | [Scaling for Training Time and Post-hoc Out-of-distribution Detection Enhancement](https://arxiv.org/abs/2111.12797) | ICLR 2024 | avail [tensorflow](docs/notebooks/tensorflow/demo_scale_tf.ipynb) or [torch](docs/notebooks/torch/demo_scale_torch.ipynb) |
 
 
 
@@ -344,6 +365,19 @@ This project received funding from the French ”Investing for the Future – PI
 
 The library was created by Paul Novello to streamline DEEL research on post-hoc deep OOD methods and foster their adoption by DEEL industrial partners. He was soon joined by Yann Pequignot, Yannick Prudent, Corentin Friedrich and Matthieu Le Goff.
 
+# Citation
+
+If you use OODEEL for your research project, please consider citing:
+```
+@misc{oodeel,
+  author = {Novello, Paul and Prudent, Yannick and Friedrich, Corentin and Pequignot, Yann and Le Goff, Matthieu},
+  title = {OODEEL, a simple, compact, and hackable post-hoc deep OOD detection for already trained tensorflow or pytorch image classifiers.},
+  year = {2023},
+  publisher = {GitHub},
+  journal = {GitHub repository},
+  howpublished = {\url{https://github.com/deel-ai/oodeel}},
+}
+```
 # License
 
 The package is released under [MIT license](LICENSE).

oodeel-0.3.0.dist-info/RECORD

@@ -0,0 +1,57 @@
+oodeel/__init__.py,sha256=-3tTNyrjeIvLCDPR0kCXup2TQX0KHn7R0emJDOKUYxI,1343
+oodeel/datasets/__init__.py,sha256=COhGN25e3TsYtIz5rJgc0aNr_k3JjGXE3cCYd5RajTQ,1398
+oodeel/datasets/data_handler.py,sha256=mWg1_Jv19rpz7p8o2DOE859J8evTRrWTiW5U_prw1sQ,10532
+oodeel/datasets/tf_data_handler.py,sha256=H3BN_N9fzT19vtJtjhe0XJG4aA-7BauGrEY4OASxc90,23102
+oodeel/datasets/torch_data_handler.py,sha256=jzQ1DvHC3xM0DOLHWGtV5OX-C63eqhabDyVDKmg11cc,24792
+oodeel/datasets/deprecated/DEPRECATED_data_handler.py,sha256=fGK3_YSbNdHuljBtrjt7xbi6ESfNpnQV_pUSn9Uot2k,7910
+oodeel/datasets/deprecated/DEPRECATED_ooddataset.py,sha256=Ad2otHV_vClK7ZY2D3-gW4QO_B3ir6DzJM1kYeNCJpw,13294
+oodeel/datasets/deprecated/DEPRECATED_tf_data_handler.py,sha256=heK9c1g829LZof8CkAmmxpTvE0EpoLZOLXuYKMrL8b4,24811
+oodeel/datasets/deprecated/DEPRECATED_torch_data_handler.py,sha256=eWCOlSPbblS8KZZAlu-M10QgY2Hv7mOq1uawLcGs0FE,26986
+oodeel/datasets/deprecated/__init__.py,sha256=VYVsYZ-_TgGyd5ls5rudytQGPocm9I7N5_M9rHcV91w,1506
+oodeel/eval/__init__.py,sha256=lQIUQjczeiRtfIqH2uLNJGubKUN6vPM93mTfY1Qz3bc,1297
+oodeel/eval/metrics.py,sha256=9fg7fVTT10_XIKrcsan6HQOlhUdFJhZ3f5VAprTKsjM,8839
+oodeel/eval/plots/__init__.py,sha256=YmcFh8RUGvljc-vCew6EiIFMDn0YA_UOfDj4eAv5_Yk,1487
+oodeel/eval/plots/features.py,sha256=fP4soXDg3BAtkBySvfUIH3Z_E3ofuU3QenH6SHRnzlM,11821
+oodeel/eval/plots/metrics.py,sha256=3QvLqEB1pAggNHeQJzUpLqL_Ro0MOJ_bPrLcLmf_qpk,4189
+oodeel/eval/plots/plotly.py,sha256=XHPFuXARrRIGVrRB7d0Od36iCn3JBP387tDrwRvPhU0,6040
+oodeel/extractor/__init__.py,sha256=Ew8gLh-xanZsWJe_gKvTa_7HciZ5yTZ-cADKuj7-PCg,1538
+oodeel/extractor/feature_extractor.py,sha256=WFNOrZjmCPt2mPr3G4ctcSqiSuvFjMZkozrwazv1zwg,6789
+oodeel/extractor/keras_feature_extractor.py,sha256=bjJfh_wgt2Y0NOGBesZ8uBjtcyiXtHYJzsWCS_tsIFg,12412
+oodeel/extractor/torch_feature_extractor.py,sha256=D_c_AV29Ciu-d_bWXk1ABad66qdX8wdddxrCgLQ-5G4,15779
+oodeel/methods/__init__.py,sha256=BkpwY0-N6SAvJ8iUxe1g0J8On-tbmLCCYTJIFI3M7i0,1721
+oodeel/methods/base.py,sha256=zeizvn6pMvdaP4FnI5TcvgrCdFARmxrwCfyU3D9LS1U,14831
+oodeel/methods/dknn.py,sha256=6eyoHAKzb3s3pkukr6r5v9_UyK8_Rj269iHQxv1PoFI,4982
+oodeel/methods/energy.py,sha256=u3TSoGDll1myCiA0FkuYZZNmGEXTTFPRSpT7-J8Q6Ec,4505
+oodeel/methods/entropy.py,sha256=dcG7oC6velSN1sky9DJAD9eD32EEPqgM4R45_iU-R18,4172
+oodeel/methods/gen.py,sha256=JafCnih09929I2KgOVaDfn8T1O4mbKdZgz3lByKvOJQ,4505
+oodeel/methods/gram.py,sha256=35yH3BYl5CgsC2fzw4GLY21MWfyLNvY0mp3tAfmJkJ4,12220
+oodeel/methods/mahalanobis.py,sha256=C6t2Gj7Ckp5KIhGFiOpk9uWY1RWIkM6h_qfLQkVGETc,7393
+oodeel/methods/mls.py,sha256=FxN1XYlTkOR2uZs07krsD8qCAfJr9UIpdKH6FYEn37Q,4399
+oodeel/methods/odin.py,sha256=Ty1YItWIsvplad72JmGuNZ_MDIBxg9iJVbjAcn33K3U,5548
+oodeel/methods/rmds.py,sha256=ji6r8_AGhIc_75-u174Mj6a8YpHC6EHDm4mFBSOsiFM,4970
+oodeel/methods/she.py,sha256=qJftwTM1UD_dtBvGH66RtejWqpkxR69ttpBy5KbNvHQ,7254
+oodeel/methods/vim.py,sha256=PMy04hMZ08THlbi66ZWg15JL9NNGBG6vndz40sez5Ps,9475
+oodeel/preprocess/__init__.py,sha256=65M9hKYHYzZ6lACA6xo7ODKofaAoqfFZ4aC5ZSJeN2I,1484
+oodeel/preprocess/tf_preprocess.py,sha256=TRaEA7KrVjWFB81vlnNk9hN-G0tGhDa97VvrBEaP9vM,4048
+oodeel/preprocess/torch_preprocess.py,sha256=BTjOyEHPfqx_CSv6Lw3zBi2wKEjAIdPN1DuBptcebA0,3976
+oodeel/types/__init__.py,sha256=9TTXjSBfbaDIVMRnclInHI-CBr4L6VZTi61rCJKcTw8,2484
+oodeel/utils/__init__.py,sha256=Lue9BysqeJf5Ej0szafgC9na8IZZR7hWGInJxoEiHUg,1696
+oodeel/utils/general_utils.py,sha256=xc6e7q19ALgMxdCgS7TIyDiMUIGF4ih-aTK1kSlqWoQ,3292
+oodeel/utils/operator.py,sha256=ETAFJ_oYhiD1Rawjooueq5KDl4SNzJR5fQDUU05uMz8,8262
+oodeel/utils/tf_operator.py,sha256=gHJZlD6SKOBxDtwv0oy6u93oXYiS5gylKv5lbVWlzx4,9385
+oodeel/utils/tf_training_tools.py,sha256=31cPCpXKmO7lMRrefmY93m6cSn0DRaHxVFeYgUiE6kI,8090
+oodeel/utils/torch_operator.py,sha256=lO_Onb4km6UNyqYjRkFPi3OcoT2MUwE1ktzP_JMDmFY,10199
+oodeel/utils/torch_training_tools.py,sha256=ggL_iDwyquTw9CtQAg--IODVdS1BFsBw4U5IOVlAsK8,11192
+oodeel-0.3.0.dist-info/licenses/LICENSE,sha256=XrlZ0uYNVeUAF-iEVX21J3CTJjYPgIZUagYSy3Hf0jk,1265
+tests/__init__.py,sha256=lQIUQjczeiRtfIqH2uLNJGubKUN6vPM93mTfY1Qz3bc,1297
+tests/tools_operator.py,sha256=YTU_FppXZPUyi3nldQsDkCMQ3Bvz1Hn_V7L45EEwftc,5328
+tests/tests_tensorflow/__init__.py,sha256=VuiDSdOBB2jUeobzAW-XTtHJQrcyZpp8i2DyFu0A2RI,1710
+tests/tests_tensorflow/tf_methods_utils.py,sha256=wHjsvKJj1EhcmAQhDgSWfRbVZR_jJwknT8vyez0YByw,5277
+tests/tests_tensorflow/tools_tf.py,sha256=Z_MEzhwCOF7E7uT-tnfRS3Po9hg0MFzbESP7sMAFBkY,3499
+tests/tests_torch/__init__.py,sha256=3mVxix2Ecn2wUo9DxvzyJBmyOAkv4fVWHHp6uLQHoic,1738
+tests/tests_torch/tools_torch.py,sha256=lS5_kctnyiB6eP-NTfVUsbj5Mn-AiRO0Jo75p8djmwg,4975
+tests/tests_torch/torch_methods_utils.py,sha256=PAxBs2smjsDtQjgwrwdbDzvVrboczc0pQOKtNfl82J4,5432
+oodeel-0.3.0.dist-info/METADATA,sha256=YNvMUC4osqrw2FWlqqNMJiapvjBGFwxp834EOkNIYtg,20299
+oodeel-0.3.0.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
+oodeel-0.3.0.dist-info/top_level.txt,sha256=zkYRty1FGJ1dkpk-5MU_4uFfBFmcxXoqSwej73xELDs,13
+oodeel-0.3.0.dist-info/RECORD,,

{oodeel-0.1.1.dist-info → oodeel-0.3.0.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: bdist_wheel (0.41.2)
+Generator: setuptools (78.1.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

tests/tests_tensorflow/tf_methods_utils.py

@@ -66,7 +66,8 @@ def load_blobs_data(batch_size=128, num_samples=10000, train_ratio=0.8):
 def load_blob_mlp():
     model_path_blobs = tf.keras.utils.get_file(
         "blobs_mlp.h5",
-        origin="https://share.deel.ai/s/bc5jx9HQAGYya9m/download/blobs_mlp.h5",
+        origin="https://github.com/deel-ai/oodeel/blob/assets/test_models/"
+        + "blobs_mlp.h5?raw=True",
         cache_dir=model_path,
         cache_subdir="",
     )