AgML 0.4.5

This feature adds a couple of new features as well as bugfixes for the existing API.

Main Changes

• You can now use image classification models (agml.models.ClassificationModel) as image regression models by passing regression = True upon instantiation. This drops the final argmax computation and returns the softmax regression values.
• Pass a custom set of RGB values to agml.viz.set_colormap to use a custom colormap.
• A new preprocessing function has been added to agml.models.preprocessing: agml.models.preprocessing.imagenet_preprocess, which prepares images for input to an ImageNet-backend model (image classification, semantic segmentation).

Bugfixes

• The MeanAveragePrecision metric has been fixed, and no longer throws errors for empty predictions (or for early-stage training results).
• Custom object detection datasets can now be auto-loaded and classes automatically inferred without throwing an error.

Read the Full Changelog Here.

AgML 0.4.4

This release introduces a number of updates and bugfixes.

Main Changes

agml.synthetic

• Helios compilation has now been optimized, meaning that canopy generation speed should be increased by 3 to 5 times.
• When using HeliosOptions.camera.generate_positions(), camera views are no longer angled by default.

agml.models

• The agml.models.ClassificationModel is now initialized with num_classes, rather than a dataset name.
• For the agml.models.SegmentationModel.predict() method, a new argument, overlay, has been added, which allows the resulting segmentation mask to be overlaid or displayed side-by-side.

agml.viz

• A new method has been added, agml.viz.convert_figure_to_image(), which converts the in-built Matplotlib figure when using agml.viz methods into an image array.

Read the Full Changelog Here.

AgML 0.4.3

This release contains additional bugfixes for Helios installation and compilation.

AgML 0.4.2

This release fixes installation problems when installing Helios.

Bugfixes

• Helios will now auto-install when running any configuration or compilation methods, rather than throwing an installation error.
• The installation script which fetches Helios is now correctly included with the package.

Note: AgML v0.4.1 contains a partial fix of these errors, so please use this version instead. This GitHub release collapses v0.4.1 and v0.4.2 into a single release.

AgML 0.4.0

This is the first release in the AgML v0.4.x cycle, introducing the agml.synthetic module backed by the Helios API (https://baileylab.ucdavis.edu/software/helios/) for generating synthetic agricultural data.

API Changes

agml.synthetic

• Introduction of the agml.synthetic module.
• Generate data using the agml.synthetic.HeliosDataGenerator.
  • Customize generated data, including task, environment, and canopy parameters, using agml.synthetic.HeliosOptions.
  • Convert data between the Helios and AgML formats automatically upon generation.
• Customize the compilation of Helios using agml.synthetic.recompile_helios.
• Get default parameters and information using agml.synthetic.available_canopies() and agml.synthetic.default_canopy_parameters().

agml.data

• Load Helios-generated synthetic datasets using agml.data.AgMLDataLoader.helios().

agml.viz

• Added new inspection methods for synthetic data:
  • agml.viz.plot_synthetic_camera_positions creates a 3D plot from camera position and lookat vectors, to understand the geometry of the environment.
  • agml.viz.vizualize_all_views plots all of the views for a specific generated canopy.

agml.backend

• Similar to the methods for public datasets and models, synthetic datasets can be generated to a specific path by using agml.backend.set_synthetic_save_path(), and retrieved using agml.backend.synthetic_save_path().

Read The Full Changelog Here

AgML 0.3.0

This is the first release in AgML v0.3.x cycle, introducing the agml.models module.

API Changes

agml.models

• Introduction of the agml.models module. To load the module, include an import agml.models statement after the standard import agml.
• Construct inference models for the following tasks:
  • image classification: agml.models.ClassificationModel (using an EfficientNetB4 model)
  • semantic segmentation: agml.models.SegmentationModel (using a DeepLabV3 model)
  • object detection: agml.models.DetectionModel (using an EfficientDetD4 model)
• Use model.predict on an input data sample to get the prediction results, or use model.show_prediction to visualize the prediction for semantic segmentation and object detection datasets.
• Use model.preprocess_input to format inputs in the required format for each model.
• Load a pretrained benchmark for an AgML public dataset using model.load_benchmark.
  • Currently available for semantic segmentation and object detection.
  • Print information about the benchmark metric and training parameter using the model.benchmark property.
• Evaluate a model using standard metrics using model.evaluate() with an AgMLDataLoader
• Standard metric containers using the agml.models.metrics module: currently with Accuracy for image classification and MeanAveragePrecision for object detection.
• Preprocessing pipelines using agml.models.preprocessing: currently with EfficientDetPreprocessor for object detection models.
• Standard loss functions with agml.models.losses: currently with DiceLoss for semantic segmentation.

agml.data

• Load multiple datasets into a single AgMLDataLoader using the new Sequence API: agml.data.AgMLDataLoader([<d1>, <d2>]).
• Load custom datasets into an AgMLDataLoader using agml.data.AgMLDataLoader.custom(<d>).
• Merge two loaders into a single AgMLDataLoader using agml.data.AgMLDataLoader.merge(<l1>, <l2>).
• Copy the state of another AgMLDataLoader using agml.data.AgMLDataLoader.copy_state(<d>). This copies the states of the internal managers, so the original loader will get the transforms, resizing, and training mode of the copied loader.
• New Dataset Reduction Methods:
  • Use loader.take_random to select a random number of data samples from the loader and return a new loader with this reduced number of samples.
  • Use loader.take_class on a multi-dataset object detection AgMLDataLoader to return a reduced loader containing all samples belonging to the specific class (or multiple classes).
  • Use loader.take_dataset on a multi-dataset AgMLDataLoader to return one of the individual loaders in the collection.
• Preprocessing Changes:
  • Select a custom method for image resizing using the method keyword argument in AgMLDataLoader.resize_images().
  • You can update whether the loader should be auto-shuffled internally through the shuffle_data property (set to True or False).
  • Semantic segmentation loaders now have an option loader.mask_to_channel_basis (the parallel of loader.labels_to_one_hot for image classification), which converts one-channel masks to multi-channel masks with each channel representing a binary mask for a class.
  • For multi-dataset object detection loaders, the loader.generalize_class_detections method generalizes all classes, thus resulting in the loader having only one-class (useful for localization-intensive tasks).
  • The agml.data.convert_bbox_format method now allows you to pass string formats, e.g., pascal-voc or efficientdet: these get auto-expanded to the full bounding box format in the method itself.

agml.backend

• Similar to the method for datasets, models can now be saved to a specific path using agml.backend.set_model_save_path() and this path can be retrieved using agml.backend.model_save_path().
• A new agml.backend.experimental module has been added for experimental features.
  • Currently, the first experimental feature is splitting already-split loaders, with agml.backend.experimental.allow_nested_data_splitting().

agml.viz

• New methods have been introduced which allow visualization of real and predicted samples:
  • Visualize real and predicted bounding boxes with agml.viz.visualize_real_and_predicted_boxes.
  • Visualize an image with multiple sets of bounding boxes using agml.viz.visualize_image_and_many_boxes.
  • Visualize a real and predicted semantic segmentation mask with agml.viz.visualize_image_mask_and_predicted.

Behavior Changes

• The AgMLDataLoader now has a set of new properties (num_images, num_classes, etc). These are similar to the properties loader.info.num_images, loader.images.num_classes, etc., however, the properties directly on the loader reflect any changes applied to reduction methods.
  • For instance, if you call loader.split(), then the reduced loader.train_data loader will have a reduced number of loader.num_images, while loader.info.num_images will continue to reflect the total number of images for the original dataset.
• Dataset transforms are now tracked by their time of insertion; if you call loader.transform() twice, the transforms which are passed in the first call will be applied before those in the second call.
• Any methods which use random operations (such as loader.take_random) now have an optional random_state keyword argument, which can be passed to control the random seed used for that method only.
• All transformation methods now have an optional argument add which allows them to be disabled.
• When converting an AgMLDataLoader into training mode, it now disables the auto mode so that no inadvertent resizing is done unless the user explicitly requests it.
• In all agml.viz methods, image arrays are auto-converted to 8-bit unsigned integer format before display. Additionally, the format_image method now takes an additional argument mask for preprocessing masks.
• Array check methods now check for NumPy arrays and lists, before PyTorch tensors, before finally TensorFlow tensors: this prevents having to load in the full API for PyTorch/TensorFlow modules when unnecessary.

Other Bugfixes

• Bounding boxes for object datasets are now clipped at [0, image_size) on loading.
• Exporting COCO datasets from a split AgMLDataLoader now returns the correct, reduced COCO JSON dictionary, rather than the entire dataset's COCO JSON.
• Copying a loader or splitting it will now update all internal states; previously there was a chance the TrainingManager would have different resize/transform managers than the upper-level DataManager.
• You can now download a list of datasets using agml.data.download_public_dataset.
• When visualizing semantic segmentation masks with no class labels (e.g., a prediction with only zeros), now only the original image is displayed, rather than throwing an error.
• An error with namedtuples in local scopes which prevented AgMLMetadata from being serialized has been resolved.

Read the Full Changelog Here.

AgML 0.2.9

This release fixes a small breaking change, updating the AWS download URL for datasets.

AgML 0.2.8

This release introduces three new semantic segmentation datasets:

1. peachpear_flower_segmentation
2. red_grapes_and_leaves_segmentation
3. white_grapes_and_leaves_segmentation

Read the Full Changelog Here.

AgML 0.2.7

This release introduces two new object detection datasets: plant_doc_detection and wheat_head_counting.

In addition, a number of fixes have been made to existing datasets:

• plant_doc_classification: Class names have been updated.
• fruit_detection_worldwide: Class indexing has been redone from 1 instead of 0 (to fit the object detection standard).
• apple_detection_usa: There was an issue where multiple images had the same ID, this has been fixed so each image has a unique ID.
• rice_seedling_segmentation: Removed the redundant background class and re-indexed from 1.

Read the Full Changelog Here.

AgML 0.2.6

This is a bugfix-oriented release. The biggest new change is the removal of the cotton_seedling_counting dataset.

Bugfixes

1. Fixed an issue regarding splitting datasets using proportions (float values).
2. Updated the number of images in plant_seedlings_aarhus to reflect the actual count.
3. Updated the error message displayed when accessing nonexistent metadata.