Feature/comet logging

README.md

@@ -14,6 +14,7 @@
 [![python](https://img.shields.io/badge/python-3.7%2B-green)]()
 [![pytorch](https://img.shields.io/badge/pytorch-1.8.1%2B-orange)]()
 [![openvino](https://img.shields.io/badge/openvino-2021.4.2-purple)]()
+[![comet](https://custom-icon-badges.herokuapp.com/badge/comet__ml-3.31.7-orange?logo=logo_comet_ml)](https://www.comet.com/site/products/ml-experiment-tracking/?utm_source=anomalib&utm_medium=referral)
 [![Codacy Badge](https://app.codacy.com/project/badge/Grade/684927c1c76c4c5e94bb53480812fbbb)](https://www.codacy.com/gh/openvinotoolkit/anomalib/dashboard?utm_source=github.com&utm_medium=referral&utm_content=openvinotoolkit/anomalib&utm_campaign=Badge_Grade)
 [![black](https://img.shields.io/badge/code%20style-black-000000.svg)]()
 [![Nightly-Regression Test](https://github.com/openvinotoolkit/anomalib/actions/workflows/nightly.yml/badge.svg)](https://github.com/openvinotoolkit/anomalib/actions/workflows/nightly.yml)
@@ -241,19 +242,25 @@ python tools/benchmarking/benchmark.py \
 
 Refer to the [Benchmarking Documentation](https://openvinotoolkit.github.io/anomalib/guides/benchmarking.html) for more details.
 
-# Logging Images
+# Experiment Management 
 
-You can save images locally or to a logger such TensorBoard or Weights and Biases by setting the following configuration.
+Anomablib is integrated with various libraries for experiment tracking such as Comet, tensorboard, and wandb through [pytorch lighting loggers](https://pytorch-lightning.readthedocs.io/en/stable/extensions/logging.html). 
+
+Below is an example of how to enable logging for hyper-parameters, metrics, model graphs, and predictions on images in the test data-set
 
 ```yaml
-logging:
-  logger: [tensorboard, wandb]
-  log_graph: false
+visualization:
+  log_images: True # log images to the available loggers (if any)
+  mode: full # options: ["full", "simple"]
+
+ logging:
+  logger: [comet, tensorboard, wandb]
+  log_graph: True
 ```
 
-For more information on logging images, refer to the [Logging Documentation](https://openvinotoolkit.github.io/anomalib/guides/logging.html)
+For more information, refer to the [Logging Documentation](https://openvinotoolkit.github.io/anomalib/guides/logging.html)
 
----
+Note: Set your API Key for [Comet.ml](https://www.comet.com/signup?utm_source=anomalib&utm_medium=referral) via `comet_ml.init()` in interactive python or simply run `export COMET_API_KEY=<Your API Key>`
 
 # Datasets
 

anomalib/models/cflow/config.yaml

@@ -59,7 +59,7 @@ project:
   path: ./results
 
 logging:
-  logger: [] # options: [tensorboard, wandb, csv] or combinations.
+  logger: [] # options: [comet, tensorboard, wandb, csv] or combinations.
   log_graph: false # Logs the model graph to respective logger.
 
 # PL Trainer Args. Don't add extra parameter here.

anomalib/models/dfkde/config.yaml

@@ -47,7 +47,7 @@ project:
   path: ./results
 
 logging:
-  logger: [] # options: [tensorboard, wandb, csv] or combinations.
+  logger: [] # options: [comet, tensorboard, wandb, csv] or combinations.
   log_graph: false # Logs the model graph to respective logger.
 
 # PL Trainer Args. Don't add extra parameter here.

anomalib/models/dfm/config.yaml

@@ -47,7 +47,7 @@ project:
   path: ./results
 
 logging:
-  logger: [] # options: [tensorboard, wandb, csv] or combinations.
+  logger: [] # options: [comet, tensorboard, wandb, csv] or combinations.
   log_graph: false # Logs the model graph to respective logger.
 
 # PL Trainer Args. Don't add extra parameter here.

anomalib/models/draem/config.yaml

@@ -56,7 +56,7 @@ project:
   path: ./results
 
 logging:
-  logger: [] # options: [tensorboard, wandb, csv] or combinations.
+  logger: [] # options: [comet, tensorboard, wandb, csv] or combinations.
   log_graph: false # Logs the model graph to respective logger.
 
 optimization:

anomalib/models/fastflow/config.yaml

@@ -59,7 +59,7 @@ project:
   path: ./results
 
 logging:
-  logger: [] # options: [tensorboard, wandb, csv] or combinations.
+  logger: [] # options: [comet, tensorboard, wandb, csv] or combinations.
   log_graph: false # Logs the model graph to respective logger.
 
 # PL Trainer Args. Don't add extra parameter here.

anomalib/models/ganomaly/config.yaml

@@ -59,7 +59,7 @@ project:
   path: ./results
 
 logging:
-  logger: [] # options: [tensorboard, wandb, csv] or combinations.
+  logger: [] # options: [comet, tensorboard, wandb, csv] or combinations.
   log_graph: false # Logs the model graph to respective logger.
 
 optimization:

anomalib/models/padim/config.yaml

@@ -54,7 +54,7 @@ project:
   path: ./results
 
 logging:
-  logger: [] # options: [tensorboard, wandb, csv] or combinations.
+  logger: [] # options: [comet, tensorboard, wandb, csv] or combinations.
   log_graph: false # Logs the model graph to respective logger.
 
 optimization:

anomalib/models/patchcore/config.yaml

@@ -55,7 +55,7 @@ project:
   path: ./results
 
 logging:
-  logger: [] # options: [tensorboard, wandb, csv] or combinations.
+  logger: [] # options: [comet, tensorboard, wandb, csv] or combinations.
   log_graph: false # Logs the model graph to respective logger.
 
 # PL Trainer Args. Don't add extra parameter here.

anomalib/models/reverse_distillation/config.yaml

@@ -63,7 +63,7 @@ project:
   path: ./results
 
 logging:
-  logger: [] # options: [tensorboard, wandb, csv] or combinations.
+  logger: [] # options: [comet, tensorboard, wandb, csv] or combinations.
   log_graph: false # Logs the model graph to respective logger.
 
 optimization:

anomalib/models/stfpm/config.yaml

@@ -61,7 +61,7 @@ project:
   path: ./results
 
 logging:
-  logger: [] # options: [tensorboard, wandb, csv] or combinations.
+  logger: [] # options: [comet, tensorboard, wandb, csv] or combinations.
   log_graph: false # Logs the model graph to respective logger.
 
 optimization:

anomalib/utils/callbacks/graph.py

@@ -7,7 +7,7 @@
 from pytorch_lightning import Callback, LightningModule, Trainer
 from pytorch_lightning.utilities.cli import CALLBACK_REGISTRY
 
-from anomalib.utils.loggers import AnomalibTensorBoardLogger, AnomalibWandbLogger
+from anomalib.utils.loggers import AnomalibTensorBoardLogger, AnomalibWandbLogger, AnomalibCometLogger
 
 
 @CALLBACK_REGISTRY
@@ -38,7 +38,7 @@ def on_train_end(self, trainer: Trainer, pl_module: LightningModule) -> None:
         """
 
         for logger in trainer.loggers:
-            if isinstance(logger, AnomalibTensorBoardLogger):
+            if isinstance(logger, (AnomalibCometLogger, AnomalibTensorBoardLogger)):
                 logger.log_graph(pl_module, input_array=torch.ones((1, 3, 256, 256)))
             elif isinstance(logger, AnomalibWandbLogger):
                 logger.unwatch(pl_module)  # type: ignore

anomalib/utils/loggers/__init__.py

@@ -14,14 +14,18 @@
 
 from .tensorboard import AnomalibTensorBoardLogger
 from .wandb import AnomalibWandbLogger
+from .comet import AnomalibCometLogger
 
 __all__ = [
+    "AnomalibCometLogger",
     "AnomalibTensorBoardLogger",
     "AnomalibWandbLogger",
     "configure_logger",
     "get_experiment_logger",
 ]
-AVAILABLE_LOGGERS = ["tensorboard", "wandb", "csv"]
+
+
+AVAILABLE_LOGGERS = ["tensorboard", "wandb", "csv", "comet"]
 
 
 logger = logging.getLogger(__name__)
@@ -112,6 +116,21 @@ def get_experiment_logger(
                     save_dir=wandb_logdir,
                 )
             )
+        elif experiment_logger=="comet":
+            comet_logdir = os.path.join(config.project.path, "logs")
+            os.makedirs(comet_logdir, exist_ok=True)
+            run_name = (
+                config.model.name
+                if "category" not in config.dataset.keys()
+                else f"{config.dataset.category} {config.model.name}"
+            )
+            logger_list.append(
+                AnomalibCometLogger(
+                    project_name=config.dataset.name,
+                    experiment_name=run_name,
+                    save_dir=comet_logdir
+                )
+            )
         elif experiment_logger == "csv":
             logger_list.append(CSVLogger(save_dir=os.path.join(config.project.path, "logs")))
         else:

anomalib/utils/loggers/comet.py

@@ -0,0 +1,117 @@
+"""comet logger with add image interface."""
+
+# Copyright (C) 2022 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+from typing import Any, Optional, Union
+
+import numpy as np
+from matplotlib.figure import Figure
+from pytorch_lightning.loggers.comet import CometLogger
+from pytorch_lightning.utilities import rank_zero_only
+
+
+from .base import ImageLoggerBase
+
+
+class AnomalibCometLogger(ImageLoggerBase, CometLogger):
+    """Logger for comet.
+
+    Adds interface for `add_image` in the logger rather than calling the experiment object.
+
+    Note:
+        Same as the CometLogger provided by PyTorch Lightning and the doc string is reproduced below.
+
+    Track your parameters, metrics, source code and more using
+    `Comet <https://www.comet.com/?utm_source=pytorch_lightning&utm_medium=referral>`_.
+
+    Install it with pip:
+
+    .. code-block:: bash
+
+        pip install comet-ml
+
+    Comet requires either an API Key (online mode) or a local directory path (offline mode).
+
+    Args:
+        api_key: Required in online mode. API key, found on Comet.ml. If not given, this
+            will be loaded from the environment variable COMET_API_KEY or ~/.comet.config
+            if either exists.
+        save_dir: Required in offline mode. The path for the directory to save local
+            comet logs. If given, this also sets the directory for saving checkpoints.
+        project_name: Optional. Send your experiment to a specific project.
+            Otherwise will be sent to Uncategorized Experiments.
+            If the project name does not already exist, Comet.ml will create a new project.
+        rest_api_key: Optional. Rest API key found in Comet.ml settings.
+            This is used to determine version number
+        experiment_name: Optional. String representing the name for this particular experiment on Comet.ml.
+        experiment_key: Optional. If set, restores from existing experiment.
+        offline: If api_key and save_dir are both given, this determines whether
+            the experiment will be in online or offline mode. This is useful if you use
+            save_dir to control the checkpoints directory and have a ~/.comet.config
+            file but still want to run offline experiments.
+        prefix: A string to put at the beginning of metric keys.
+        \**kwargs: Additional arguments like `workspace`, `log_code`, etc. used by
+            :class:`CometExperiment` can be passed as keyword arguments in this logger.
+
+    Raises:
+        ModuleNotFoundError:
+            If required Comet package is not installed on the device.
+        MisconfigurationException:
+            If neither ``api_key`` nor ``save_dir`` are passed as arguments.
+
+    Example:
+        >>> from anomalib.utils.loggers import AnomalibCometLogger
+        >>> from pytorch_lightning import Trainer
+        >>> comet_logger = AnomalibCometLogger()
+        >>> trainer = Trainer(logger=comet_logger)
+
+    See Also:
+        - `Demo in Google Colab <https://colab.research.google.com/github/comet-ml/comet-examples/blob/master/integrations/model-training/pytorch-lightning/notebooks/Comet_and_Pytorch_Lightning.ipynb>`__
+        - `Comet Documentation <https://www.comet.com/docs/v2/integrations/ml-frameworks/pytorch-lightning/>`__
+    """
+    def __init__(
+        self,
+        api_key: Optional[str] = None,
+        save_dir: Optional[str] = None,
+        project_name: Optional[str] = None,
+        rest_api_key: Optional[str] = None,
+        experiment_name: Optional[str] = None,
+        experiment_key: Optional[str] = None,
+        offline: bool = False,
+        prefix: str = "",
+        **kwargs
+    ) -> None:
+        super().__init__(
+            api_key=api_key,
+            save_dir=save_dir,
+            project_name=project_name,
+            rest_api_key=rest_api_key,
+            experiment_name=experiment_name,
+            experiment_key=experiment_key,
+            offline=offline,
+            prefix=prefix,
+            **kwargs
+        )
+        self.experiment.log_other("Created from", "Anomalib")
+
+
+    @rank_zero_only
+    def add_image(self, image: Union[np.ndarray, Figure], name: Optional[str] = None, **kwargs: Any):
+        """Interface to add image to comet logger.
+
+        Args:
+            image (Union[np.ndarray, Figure]): Image to log
+            name (Optional[str]): The tag of the image
+            kwargs: Accepts only `global_step` (int). The step at which to log the image.
+        """
+        if "global_step" not in kwargs:
+            raise ValueError("`global_step` is required for comet logger")
+
+        global_step = kwargs["global_step"]
+        # Need to call different functions of `Experiment` for  Figure vs np.ndarray
+
+        if isinstance(image, Figure):
+            self.experiment.log_figure(figure_name=name, figure=image, step=global_step)
+        else:
+            self.experiment.log_image(name=name, image_data=image, step=global_step)

docs/blog/001-train-custom-dataset/README.md

@@ -151,11 +151,11 @@ Here is an example of the generated results for a toy dataset containing Hazelnu
 
 ## Logging and Experiment Management
 
-While it is delightful to know how good your model performed on your preferred metric, it is even more exciting to see the predicted outputs. Anomalib provides a couple of ways to log and track experiments. These can be used individually or in a combination. As of the current release, you can save images to a local folder, or upload to weights and biases, or TensorBoard.
+While it is delightful to know how good your model performed on your preferred metric, it is even more exciting to see the predicted outputs. Anomalib provides a couple of ways to log and track experiments. These can be used individually or in a combination. As of the current release, you can save images to a local folder, or upload to comet, weights and biases, or TensorBoard.
 
-To select where you would like to save the images, change the `log_images_to` parameter in the `project` section in the config file.
+To select where you would like to save the images, change the `log_images` parameter in the `Visualization` section in the config file to true.
 
-For example, setting the following `log_images_to: ["local"]` will result in saving the images in the results folder as shown in the tree structure below:
+For example, setting the following `log_images: True` will result in saving the images in the results folder as shown in the tree structure below:
 
 ```bash
 results
@@ -176,14 +176,21 @@ results
 
 ### Logging to Tensorboard and/or W&B
 
-To use TensorBoard and/or W&B logger, ensure that the logger parameter is set to `tensorboard`, `wandb` or `[tensorboard, wandb]` in the `logging` section of the config file.
+To use TensorBoard and/or W&B logger and/or Comet logger, ensure that the logger parameter is set to `comet`, `tensorboard`, `wandb` or `[tensorboard, wandb]` in the `logging` section of the config file.
 
-An example configuration for saving to TensorBoard is shown in the figure below. Similarly after setting logger to `wandb` you will see the images on your wandb project dashboard.
+An example configuration for saving to TensorBoard is shown in the figure below. Similarly after setting logger to `wandb` or 'comet' you will see the images on your wandb and/or comet project dashboard.
 
 ```yaml
-logging:
-  log_images_to: [tensorboard]
-  logger: tensorboard # options: [tensorboard, wandb, csv] or combinations.
+  visualization:
+    show_images: False # show images on the screen
+    save_images: False # save images to the file system
+    log_images: True # log images to the available loggers (if any)
+    image_save_path: null # path to which images will be saved
+    mode: full # options: ["full", "simple"]
+
+  logging:
+    logger: [comet, tensorboard, wandb] #Choose any combination of these 3
+    log_graph: false
 ```
 
 <div align="center">