📝 Documentation refactor (#576)

* Add benchmark to tutorial

* Move export to tutorials

* Move hpo to tutorials

* Move inference to tutorials

* Move logging to tutorials

* Create installation in tutorials

* Create training to tutorials

* Create tutorials index

* Update conf.py file

* Add anomalib logos to logos directory

* Add data docs

* Add algos

* Add model docs

* Add reference api

* Remove blank line in metrics

* Add reference guide

* Add how to guides

* Add developer guide

* Add blog to how-to-guide

* Remove guides directory

* Add train custom data to how-to-guides

* Fix typos

* Add notebooks to how-to-guides

* Add anomalib favicon

* Add missing algo descriptions

* Rename Reference to Reference Guide

* Add how to add a new model

* fix typos

* Merge PR 544

* Minor refactor (#587)

* 🛠 Fix PatchCore image-level score computation (#580)

* fix patchcore image-level score computation

* docstring and comment

* remove default value for n_neighbors

* torch.Tensor -> Tensor

* Minor refactor

Co-authored-by: Dick Ameln <[email protected]>
Co-authored-by: Ashwin Vaidya <[email protected]>

* Address Dicks comments

Co-authored-by: Ashwin Vaidya <[email protected]>
Co-authored-by: Dick Ameln <[email protected]>
Co-authored-by: Ashwin Vaidya <[email protected]>

Author: 4 people

README.md

@@ -117,15 +117,38 @@ where the currently available models are:
 - [STFPM](anomalib/models/stfpm)
 - [GANomaly](anomalib/models/ganomaly)
 
-## Exporting Model to ONNX or OpenVINO IR
+## Feature extraction & (pre-trained) backbones
 
-It is possible to export your model to ONNX or OpenVINO IR
+The pre-trained backbones come from [PyTorch Image Models (timm)](https://github.com/rwightman/pytorch-image-models), which are wrapped by `FeatureExtractor`.
 
-If you want to export your PyTorch model to an OpenVINO model, ensure that `export_mode` is set to `"openvino"` in the respective model `config.yaml`.
+For more information, please check our documentation or the [section about feature extraction in "Getting Started with PyTorch Image Models (timm): A Practitioner’s Guide"](https://towardsdatascience.com/getting-started-with-pytorch-image-models-timm-a-practitioners-guide-4e77b4bf9055#b83b:~:text=ready%20to%20train!-,Feature%20Extraction,-timm%20models%20also>).
+
+Tips:
+
+- Papers With Code has an interface to easily browse models available in timm: [https://paperswithcode.com/lib/timm](https://paperswithcode.com/lib/timm)
+
+- You can also find them with the function `timm.list_models("resnet*", pretrained=True)`
+
+The backbone can be set in the config file, two examples below.
+
+Anomalib < v.0.4.0
 
 ```yaml
-optimization:
-  export_mode: "openvino" # options: openvino, onnx
+model:
+  name: cflow
+  backbone: wide_resnet50_2
+  pre_trained: true
+Anomalib > v.0.4.0 Beta - Subject to Change
+```
+
+Anomalib >= v.0.4.0
+
+```yaml
+model:
+  class_path: anomalib.models.Cflow
+  init_args:
+    backbone: wide_resnet50_2
+    pre_trained: true
 ```
 
 ## Custom Dataset
@@ -222,6 +245,17 @@ python tools/inference/gradio_inference.py \
         --weights ./results/padim/mvtec/bottle/weights/model.ckpt
 ```
 
+## Exporting Model to ONNX or OpenVINO IR
+
+It is possible to export your model to ONNX or OpenVINO IR
+
+If you want to export your PyTorch model to an OpenVINO model, ensure that `export_mode` is set to `"openvino"` in the respective model `config.yaml`.
+
+```yaml
+optimization:
+  export_mode: "openvino" # options: openvino, onnx
+```
+
 # Hyperparameter Optimization
 
 To run hyperparameter optimization, use the following command:

anomalib/data/btech.py

@@ -290,7 +290,7 @@ def __init__(
             seed: seed used for the random subset splitting
             create_validation_set: Create a validation subset in addition to the train and test subsets
 
-        Examples
+        Examples:
             >>> from anomalib.data import BTech
             >>> datamodule = BTech(
             ...     root="./datasets/BTech",

anomalib/data/folder.py

@@ -350,6 +350,7 @@ def __init__(
 
         Examples:
             Assume that we use Folder Dataset for the MVTec/bottle/broken_large category. We would do:
+
             >>> from anomalib.data import Folder
             >>> datamodule = Folder(
             ...     root="./datasets/MVTec/bottle/test",
@@ -370,6 +371,7 @@ def __init__(
             The dataset expects that mask annotation filenames must be same as the original filename.
             To this end, we modified mask filenames in MVTec AD bottle category.
             Now we could try folder data module using the mvtec bottle broken large category
+
             >>> datamodule = Folder(
             ...     root="./datasets/bottle/test",
             ...     normal="good",

anomalib/data/mvtec.py

@@ -89,7 +89,7 @@ def make_mvtec_dataset(
             MVTec AD dataset does not contain a validation set. Those wanting to create a validation set
             could set this flag to ``True``.
 
-    Example:
+    Examples:
         The following example shows how to get training samples from MVTec AD bottle category:
 
         >>> root = Path('./MVTec')
@@ -313,7 +313,7 @@ def __init__(
             seed: seed used for the random subset splitting
             create_validation_set: Create a validation subset in addition to the train and test subsets
 
-        Examples
+        Examples:
             >>> from anomalib.data import MVTec
             >>> datamodule = MVTec(
             ...     root="./datasets/MVTec",