Configuration API Reference¶

This contains the API used for configuring various components of Raster Vision pipelines. This serves as the lower-level companion to the discussion of Pipelines and Commands.

rastervision.pipeline¶

class rastervision.pipeline.pipeline_config.PipelineConfig¶

Base class for configuring Pipelines.

This should be subclassed to configure new Pipelines.

root_uri¶

The root URI for output generated by the pipeline. Defaults to None.

Type: Optional[str]

rv_config¶

Used to store serialized RVConfig so pipeline can run in remote environment with the local RVConfig. This should not be set explicitly by users – it is only used by the runner when running a remote pipeline. Defaults to None.

Type: Optional[dict]

plugin_versions¶

Used to store a mapping of plugin module paths to the latest version number. This should not be set explicitly by users – it is set automatically when serializing and saving the config to disk. Defaults to None.

Type: Optional[Mapping[str, int]]

rastervision.core¶

StatsAnalyzerConfig¶

class rastervision.core.analyzer.StatsAnalyzerConfig¶

Config for an Analyzer that computes imagery statistics of scenes.

output_uri¶

URI for output. If None and this is part of an RVPipeline, this is auto-generated. Defaults to None.

Type: Optional[str]

sample_prob¶

The probability of using a random window for computing statistics. If None, will use a sliding window. Defaults to 0.1.

Type: Optional[float]

ClassConfig¶

class rastervision.core.data.ClassConfig¶

Configures the class names that are being predicted.

names¶

Names of classes.

Type: List[str]

colors¶

Colors used to visualize classes. Can be color strings accepted by matplotlib or RGB tuples. If None, a random color will be auto-generated for each class. Defaults to None.

Type: Optional[List[Union[List, str]]]

null_class¶

Optional name of class in names to use as the null class. This is used in semantic segmentation to represent the label for imagery pixels that are NODATA or that are missing a label. If None, and this Config is part of a SemanticSegmentationConfig, a null class will be added automatically. Defaults to None.

Type: Optional[str]

DatasetConfig¶

class rastervision.core.data.DatasetConfig¶

Config for a Dataset comprising the scenes for train, valid, and test splits.

class_config¶

Type: ClassConfig

train_scenes¶

Type: List[SceneConfig]

validation_scenes¶

Type: List[SceneConfig]

test_scenes¶

Defaults to [].

Type: List[SceneConfig]

SceneConfig¶

class rastervision.core.data.SceneConfig¶

Config for a Scene which comprises the raster data and labels for an AOI.

id¶

Type: str

raster_source¶

Type: RasterSourceConfig

label_source¶

Type: LabelSourceConfig

label_store¶

Defaults to None.

Type: Optional[LabelStoreConfig]

aoi_uris¶

List of URIs of GeoJSON files that define the AOIs for the scene. Each polygondefines an AOI which is a piece of the scene that is assumed to be fully labeled and usable for training or validation. Defaults to None.

Type: Optional[List[str]]

ChipClassificationLabelSourceConfig¶

class rastervision.core.data.label_source.ChipClassificationLabelSourceConfig¶

Config for a source of labels for chip classification.

This can be provided explicitly as a grid of cells, or a grid of cells can be inferred from arbitrary polygons.

vector_source¶

Type: VectorSourceConfig

ioa_thresh¶

Minimum IOA of a polygon and cell for that polygon to be a candidate for setting the class_id. Defaults to None.

Type: Optional[float]

use_intersection_over_cell¶

If True, then use the area of the cell as the denominator in the IOA. Otherwise, use the area of the polygon. Defaults to False.

Type: bool

pick_min_class_id¶

If True, the class_id for a cell is the minimum class_id of the boxes in that cell. Otherwise, pick the class_id of the box covering the greatest area. Defaults to False.

Type: bool

background_class_id¶

If not None, class_id to use as the background class; ie. the one that is used when a window contains no boxes. If not set, empty windows have None set as their class_id which is considered a null value. Defaults to None.

Type: Optional[int]

infer_cells¶

If True, infers a grid of cells based on the cell_sz. Defaults to False.

Type: bool

cell_sz¶

Size of a cell to use in pixels. If None, and this Config is part of an RVPipeline, this field will be set from RVPipeline.train_chip_sz. Defaults to None.

Type: Optional[int]

SemanticSegmentationLabelSourceConfig¶

class rastervision.core.data.label_source.SemanticSegmentationLabelSourceConfig¶

Config for a read-only label source for semantic segmentation.

raster_source¶

The labels in the form of rasters.

Type: Union[None.RasterSourceConfig, None.RasterizedSourceConfig]

rgb_class_config¶

If set, will infer the class_ids for the labels using the colors field. This assumes the labels are stored as RGB rasters. Defaults to None.

Type: Optional[ClassConfig]

ObjectDetectionLabelSourceConfig¶

class rastervision.core.data.label_source.ObjectDetectionLabelSourceConfig¶

Config for a read-only label source for object detection.

vector_source¶

Type: VectorSourceConfig

ChipClassificationGeoJSONStoreConfig¶

class rastervision.core.data.label_store.ChipClassificationGeoJSONStoreConfig¶

Config for storage for chip classification predictions.

uri¶

URI of GeoJSON file with predictions. If None, and this Config is part of a SceneConfig inside an RVPipelineConfig, it will be auto-generated. Defaults to None.

Type: Optional[str]

PolygonVectorOutputConfig¶

class rastervision.core.data.label_store.PolygonVectorOutputConfig¶

Config for vectorized semantic segmentation predictions.

uri¶

URI of vector output. If None, and this Config is part of a SceneConfig and RVPipeline, this field will be auto-generated. Defaults to None.

Type: Optional[str]

class_id¶

The prediction class that is to turned into vectors.

Type: int

denoise¶

Radius of the structural element used to remove high-frequency signals from the image. Defaults to 0.

Type: int

BuildingVectorOutputConfig¶

class rastervision.core.data.label_store.BuildingVectorOutputConfig¶

Config for vectorized semantic segmentation predictions.

Intended to break up clusters of buildings.

uri¶

URI of vector output. If None, and this Config is part of a SceneConfig and RVPipeline, this field will be auto-generated. Defaults to None.

Type: Optional[str]

class_id¶

The prediction class that is to turned into vectors.

Type: int

denoise¶

Radius of the structural element used to remove high-frequency signals from the image. Defaults to 0.

Type: int

min_aspect_ratio¶

Ratio between length and height (or height and length) of anything that can be considered to be a cluster of buildings. The goal is to distinguish between rows of buildings and (say) a single building. Defaults to 1.618.

Type: float

min_area¶

Minimum area of anything that can be considered to be a cluster of buildings. The goal is to distinguish between buildings and artifacts. Defaults to 0.0.

Type: float

element_width_factor¶

Width of the structural element used to break building clusters as a fraction of the width of the cluster. Defaults to 0.5.

Type: float

element_thickness¶

Thickness of the structural element that is used to break building clusters. Defaults to 0.001.

Type: float

SemanticSegmentationLabelStoreConfig¶

class rastervision.core.data.label_store.SemanticSegmentationLabelStoreConfig¶

Config for storage for semantic segmentation predictions.

Stores class raster as GeoTIFF, and can optionally vectorizes predictions and stores them in GeoJSON files.

uri¶

URI of file with predictions. If None, and this Config is part of a SceneConfig inside an RVPipelineConfig, this fiend will be auto-generated. Defaults to None.

Type: Optional[str]

vector_output¶

Defaults to [].

Type: List[VectorOutputConfig]

rgb¶

If True, save prediction class_ids in RGB format using the colors in class_config. Defaults to False.

Type: bool

ObjectDetectionGeoJSONStoreConfig¶

class rastervision.core.data.label_store.ObjectDetectionGeoJSONStoreConfig¶

Config for storage for object detection predictions.

uri¶

URI of GeoJSON file with predictions. If None, and this Config is part of a SceneConfig inside an RVPipelineConfig, it will be auto-generated. Defaults to None.

Type: Optional[str]

RasterioSourceConfig¶

class rastervision.core.data.raster_source.RasterioSourceConfig¶

channel_order¶

The sequence of channel indices to use when reading imagery. Defaults to None.

Type: Optional[List[int]]

transformers¶

Defaults to [].

Type: List[RasterTransformerConfig]

uris¶

List of image URIs that comprise imagery for a scene. The format of each file can be any that can be read by Rasterio/GDAL. If > 1 URI is provided, a VRT will be created to mosaic together the individual images.

Type: List[str]

x_shift¶

Defaults to 0.0.

Type: float

y_shift¶

Defaults to 0.0.

Type: float

RasterizerConfig¶

class rastervision.core.data.raster_source.RasterizerConfig¶

background_class_id¶

The class_id to use for any background pixels, ie. pixels not covered by a polygon.

Type: int

all_touched¶

If True, all pixels touched by geometries will be burned in. If false, only pixels whose center is within the polygon or that are selected by Bresenham’s line algorithm will be burned in. (See rasterio.features.rasterize). Defaults to False.

Type: bool

RasterizedSourceConfig¶

class rastervision.core.data.raster_source.RasterizedSourceConfig¶

vector_source¶

Type: VectorSourceConfig

rasterizer_config¶

Type: RasterizerConfig

StatsTransformerConfig¶

class rastervision.core.data.raster_transformer.StatsTransformerConfig¶

stats_uri¶

The URI of the output of the StatsAnalyzer. If None, and this Config is inside an RVPipeline, then this field will be auto-generated. Defaults to None.

Type: Optional[str]

VectorSourceConfig¶

class rastervision.core.data.vector_source.VectorSourceConfig¶

default_class_id¶

The default class_id to use if class cannot be inferred using other mechanisms. If a feature has an inferred class_id of None, then it will be deleted.

Type: Optional[int]

class_id_to_filter¶

Map from class_id to JSON filter used to infer missing class_ids. Each key should be a class id, and its value should be a boolean expression which is run against the property field for each feature.This allows matching different features to different class ids based on its properties. The expression schema is that described by https://docs.mapbox.com/mapbox-gl-js/style-spec/other/#other-filter. Defaults to None.

Type: Optional[Dict]

line_bufs¶

This is useful, for example, for buffering lines representing roads so that their width roughly matches the width of roads in the imagery. If None, uses default buffer value of 1. Otherwise, a map from class_id to number of pixels to buffer by. If the buffer value is None, then no buffering will be performed and the LineString or Point won’t get converted to a Polygon. Not converting to Polygon is incompatible with the currently available LabelSources, but may be useful in the future. Defaults to None.

Type: Optional[Mapping[int, Union[int, float, NoneType]]]

point_bufs¶

Same as above, but used for buffering Points into Polygons. Defaults to None.

Type: Optional[Mapping[int, Union[int, float, NoneType]]]

GeoJSONVectorSourceConfig¶

class rastervision.core.data.vector_source.GeoJSONVectorSourceConfig¶

default_class_id¶

The default class_id to use if class cannot be inferred using other mechanisms. If a feature has an inferred class_id of None, then it will be deleted.

Type: Optional[int]

class_id_to_filter¶

Map from class_id to JSON filter used to infer missing class_ids. Each key should be a class id, and its value should be a boolean expression which is run against the property field for each feature.This allows matching different features to different class ids based on its properties. The expression schema is that described by https://docs.mapbox.com/mapbox-gl-js/style-spec/other/#other-filter. Defaults to None.

Type: Optional[Dict]

line_bufs¶

This is useful, for example, for buffering lines representing roads so that their width roughly matches the width of roads in the imagery. If None, uses default buffer value of 1. Otherwise, a map from class_id to number of pixels to buffer by. If the buffer value is None, then no buffering will be performed and the LineString or Point won’t get converted to a Polygon. Not converting to Polygon is incompatible with the currently available LabelSources, but may be useful in the future. Defaults to None.

Type: Optional[Mapping[int, Union[int, float, NoneType]]]

point_bufs¶

Same as above, but used for buffering Points into Polygons. Defaults to None.

Type: Optional[Mapping[int, Union[int, float, NoneType]]]

uri¶

The URI of a GeoJSON file.

Type: str

ignore_crs_field¶

Defaults to False.

Type: bool

ChipClassificationEvaluatorConfig¶

class rastervision.core.evaluation.ChipClassificationEvaluatorConfig¶

output_uri¶

URI of JSON output by evaluator. If None, and this Config is part of an RVPipeline, then this field will be auto-generated. Defaults to None.

Type: Optional[str]

SemanticSegmentationEvaluatorConfig¶

class rastervision.core.evaluation.SemanticSegmentationEvaluatorConfig¶

output_uri¶

URI of JSON output by evaluator. If None, and this Config is part of an RVPipeline, then this field will be auto-generated. Defaults to None.

Type: Optional[str]

vector_output_uri¶

URI of evaluation of vector output. If None, and this Config is part of an RVPipeline, then this field will be auto-generated. Defaults to None.

Type: Optional[str]

ObjectDetectionEvaluatorConfig¶

class rastervision.core.evaluation.ObjectDetectionEvaluatorConfig¶

output_uri¶

URI of JSON output by evaluator. If None, and this Config is part of an RVPipeline, then this field will be auto-generated. Defaults to None.

Type: Optional[str]

ChipClassificationConfig¶

class rastervision.core.rv_pipeline.ChipClassificationConfig¶

root_uri¶

The root URI for output generated by the pipeline. Defaults to None.

Type: Optional[str]

rv_config¶

Used to store serialized RVConfig so pipeline can run in remote environment with the local RVConfig. This should not be set explicitly by users – it is only used by the runner when running a remote pipeline. Defaults to None.

Type: Optional[dict]

plugin_versions¶

Used to store a mapping of plugin module paths to the latest version number. This should not be set explicitly by users – it is set automatically when serializing and saving the config to disk. Defaults to None.

Type: Optional[Mapping[str, int]]

dataset¶

Dataset containing train, validation, and optional test scenes.

Type: DatasetConfig

backend¶

Backend to use for interfacing with ML library.

Type: BackendConfig

evaluators¶

Evaluators to run during analyzer command. If list is empty the default evaluator is added. Defaults to [].

Type: List[EvaluatorConfig]

analyzers¶

Analyzers to run during analyzer command. A StatsAnalyzer will be added automatically if any scenes have a RasterTransformer. Defaults to [].

Type: List[AnalyzerConfig]

train_chip_sz¶

Size of training chips in pixels. Defaults to 300.

Type: int

predict_chip_sz¶

Size of predictions chips in pixels. Defaults to 300.

Type: int

predict_batch_sz¶

Batch size to use during prediction. Defaults to 8.

Type: int

analyze_uri¶

URI for output of analyze. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

chip_uri¶

URI for output of chip. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

train_uri¶

URI for output of train. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

predict_uri¶

URI for output of predict. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

eval_uri¶

URI for output of eval. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

bundle_uri¶

URI for output of bundle. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

SemanticSegmentationWindowMethod¶

class rastervision.core.rv_pipeline.SemanticSegmentationWindowMethod¶

Enum for window methods

sliding¶: use a sliding window

random_sample¶: randomly sample windows

SemanticSegmentationChipOptions¶

class rastervision.core.rv_pipeline.SemanticSegmentationChipOptions¶

Chipping options for semantic segmentation.

window_method¶

Window method to use for chipping. Defaults to <SemanticSegmentationWindowMethod.sliding: ‘sliding’>.

Type: enum

target_class_ids¶

List of class ids considered as targets (ie. those to prioritize when creating chips) which is only used in conjunction with the target_count_threshold and negative_survival_probability options. Applies to the random_sample window method. Defaults to None.

Type: Optional[List[int]]

negative_survival_prob¶

List of class ids considered as targets (ie. those to prioritize when creating chips) which is only used in conjunction with the target_count_threshold and negative_survival_probability options. Applies to the random_sample window method. Defaults to 1.0.

Type: float

chips_per_scene¶

Number of chips to generate per scene. Applies to the random_sample window method. Defaults to 1000.

Type: int

target_count_threshold¶

Minimum number of pixels covering target_classes that a chip must have. Applies to the random_sample window method. Defaults to 1000.

Type: int

stride¶

Stride of windows across image. Defaults to half the chip size. Applies to the sliding_window method. Defaults to None.

Type: Optional[int]

SemanticSegmentationConfig¶

class rastervision.core.rv_pipeline.SemanticSegmentationConfig¶

root_uri¶

The root URI for output generated by the pipeline. Defaults to None.

Type: Optional[str]

rv_config¶

Used to store serialized RVConfig so pipeline can run in remote environment with the local RVConfig. This should not be set explicitly by users – it is only used by the runner when running a remote pipeline. Defaults to None.

Type: Optional[dict]

plugin_versions¶

Used to store a mapping of plugin module paths to the latest version number. This should not be set explicitly by users – it is set automatically when serializing and saving the config to disk. Defaults to None.

Type: Optional[Mapping[str, int]]

dataset¶

Dataset containing train, validation, and optional test scenes.

Type: DatasetConfig

backend¶

Backend to use for interfacing with ML library.

Type: BackendConfig

evaluators¶

Evaluators to run during analyzer command. If list is empty the default evaluator is added. Defaults to [].

Type: List[EvaluatorConfig]

analyzers¶

Analyzers to run during analyzer command. A StatsAnalyzer will be added automatically if any scenes have a RasterTransformer. Defaults to [].

Type: List[AnalyzerConfig]

train_chip_sz¶

Size of training chips in pixels. Defaults to 300.

Type: int

predict_chip_sz¶

Size of predictions chips in pixels. Defaults to 300.

Type: int

predict_batch_sz¶

Batch size to use during prediction. Defaults to 8.

Type: int

analyze_uri¶

URI for output of analyze. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

chip_uri¶

URI for output of chip. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

train_uri¶

URI for output of train. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

predict_uri¶

URI for output of predict. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

eval_uri¶

URI for output of eval. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

bundle_uri¶

URI for output of bundle. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

chip_options¶

Defaults to SemanticSegmentationChipOptions(window_method=<SemanticSegmentationWindowMethod.sliding: ‘sliding’>, target_class_ids=None, negative_survival_prob=1.0, chips_per_scene=1000, target_count_threshold=1000, stride=None, type_hint=’semantic_segmentation_chip_options’).

Type: SemanticSegmentationChipOptions

ObjectDetectionWindowMethod¶

class rastervision.core.rv_pipeline.ObjectDetectionWindowMethod¶

Enum for window methods

chip¶: the default method

ObjectDetectionChipOptions¶

class rastervision.core.rv_pipeline.ObjectDetectionChipOptions¶

neg_ratio¶

The ratio of negative chips (those containing no bounding boxes) to positive chips. This can be useful if the statistics of the background is different in positive chips. For example, in car detection, the positive chips will always contain roads, but no examples of rooftops since cars tend to not be near rooftops. Defaults to 1.0.

Type: float

ioa_thresh¶

When a box is partially outside of a training chip, it is not clear if (a clipped version) of the box should be included in the chip. If the IOA (intersection over area) of the box with the chip is greater than ioa_thresh, it is included in the chip. Defaults to 0.8.

Type: float

window_method¶

Defaults to <ObjectDetectionWindowMethod.chip: ‘chip’>.

Type: enum

label_buffer¶

Defaults to None.

Type: Optional[int]

ObjectDetectionPredictOptions¶

class rastervision.core.rv_pipeline.ObjectDetectionPredictOptions¶

merge_thresh¶

If predicted boxes have an IOA (intersection over area) greater than merge_thresh, then they are merged into a single box during postprocessing. This is needed since the sliding window approach results in some false duplicates. Defaults to 0.5.

Type: float

score_thresh¶

Predicted boxes are only output if their score is above score_thresh. Defaults to 0.5.

Type: float

ObjectDetectionConfig¶

class rastervision.core.rv_pipeline.ObjectDetectionConfig¶

root_uri¶

The root URI for output generated by the pipeline. Defaults to None.

Type: Optional[str]

rv_config¶

Used to store serialized RVConfig so pipeline can run in remote environment with the local RVConfig. This should not be set explicitly by users – it is only used by the runner when running a remote pipeline. Defaults to None.

Type: Optional[dict]

plugin_versions¶

Used to store a mapping of plugin module paths to the latest version number. This should not be set explicitly by users – it is set automatically when serializing and saving the config to disk. Defaults to None.

Type: Optional[Mapping[str, int]]

dataset¶

Dataset containing train, validation, and optional test scenes.

Type: DatasetConfig

backend¶

Backend to use for interfacing with ML library.

Type: BackendConfig

evaluators¶

Evaluators to run during analyzer command. If list is empty the default evaluator is added. Defaults to [].

Type: List[EvaluatorConfig]

analyzers¶

Analyzers to run during analyzer command. A StatsAnalyzer will be added automatically if any scenes have a RasterTransformer. Defaults to [].

Type: List[AnalyzerConfig]

train_chip_sz¶

Size of training chips in pixels. Defaults to 300.

Type: int

predict_chip_sz¶

Size of predictions chips in pixels. Defaults to 300.

Type: int

predict_batch_sz¶

Batch size to use during prediction. Defaults to 8.

Type: int

analyze_uri¶

URI for output of analyze. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

chip_uri¶

URI for output of chip. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

train_uri¶

URI for output of train. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

predict_uri¶

URI for output of predict. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

eval_uri¶

URI for output of eval. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

bundle_uri¶

URI for output of bundle. If None, will be auto-generated. Defaults to None.

Type: Optional[str]

chip_options¶

Defaults to ObjectDetectionChipOptions(neg_ratio=1.0, ioa_thresh=0.8, window_method=<ObjectDetectionWindowMethod.chip: ‘chip’>, label_buffer=None, type_hint=’object_detection_chip_options’).

Type: ObjectDetectionChipOptions

predict_options¶

Defaults to ObjectDetectionPredictOptions(merge_thresh=0.5, score_thresh=0.5, type_hint=’object_detection_predict_options’).

Type: ObjectDetectionPredictOptions

rastervision.pytorch_backend¶

PyTorchChipClassificationConfig¶

class rastervision.pytorch_backend.PyTorchChipClassificationConfig¶

model¶

Type: ClassificationModelConfig

solver¶

Type: SolverConfig

log_tensorboard¶

If True, log events to Tensorboard log files. Defaults to True.

Type: bool

run_tensorboard¶

If True, run Tensorboard server pointing at log files. Defaults to False.

Type: bool

augmentors¶

Names of albumentations augmentors to use for training batches. Choices include: [‘Blur’, ‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’, ‘GaussianBlur’, ‘GaussNoise’, ‘RGBShift’, ‘ToGray’]. Defaults to [‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’].

Type: List[str]

test_mode¶

This field is passed along to the LearnerConfig which is returned by get_learner_config(). For more info, see the docs forpytorch_learner.learner_config.LearnerConfig.test_mode. Defaults to False.

Type: bool

PyTorchSemanticSegmentationConfig¶

class rastervision.pytorch_backend.PyTorchSemanticSegmentationConfig¶

model¶

Type: SemanticSegmentationModelConfig

solver¶

Type: SolverConfig

log_tensorboard¶

If True, log events to Tensorboard log files. Defaults to True.

Type: bool

run_tensorboard¶

If True, run Tensorboard server pointing at log files. Defaults to False.

Type: bool

augmentors¶

Names of albumentations augmentors to use for training batches. Choices include: [‘Blur’, ‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’, ‘GaussianBlur’, ‘GaussNoise’, ‘RGBShift’, ‘ToGray’]. Defaults to [‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’].

Type: List[str]

test_mode¶

This field is passed along to the LearnerConfig which is returned by get_learner_config(). For more info, see the docs forpytorch_learner.learner_config.LearnerConfig.test_mode. Defaults to False.

Type: bool

PyTorchObjectDetectionConfig¶

class rastervision.pytorch_backend.PyTorchObjectDetectionConfig¶

model¶

Type: ObjectDetectionModelConfig

solver¶

Type: SolverConfig

log_tensorboard¶

If True, log events to Tensorboard log files. Defaults to True.

Type: bool

run_tensorboard¶

If True, run Tensorboard server pointing at log files. Defaults to False.

Type: bool

augmentors¶

Names of albumentations augmentors to use for training batches. Choices include: [‘Blur’, ‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’, ‘GaussianBlur’, ‘GaussNoise’, ‘RGBShift’, ‘ToGray’]. Defaults to [‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’].

Type: List[str]

test_mode¶

This field is passed along to the LearnerConfig which is returned by get_learner_config(). For more info, see the docs forpytorch_learner.learner_config.LearnerConfig.test_mode. Defaults to False.

Type: bool

rastervision.pytorch_learner¶

Backbone¶

class rastervision.pytorch_learner.Backbone¶

An enumeration.

alexnet = 'alexnet'¶

densenet121 = 'densenet121'¶

densenet161 = 'densenet161'¶

densenet169 = 'densenet169'¶

densenet201 = 'densenet201'¶

googlenet = 'googlenet'¶

inception_v3 = 'inception_v3'¶

int_to_str = <function Backbone.int_to_str>¶

mnasnet0_5 = 'mnasnet0_5'¶

mnasnet0_75 = 'mnasnet0_75'¶

mnasnet1_0 = 'mnasnet1_0'¶

mnasnet1_3 = 'mnasnet1_3'¶

mobilenet_v2 = 'mobilenet_v2'¶

resnet101 = 'resnet101'¶

resnet152 = 'resnet152'¶

resnet18 = 'resnet18'¶

resnet34 = 'resnet34'¶

resnet50 = 'resnet50'¶

resnext101_32x8d = 'resnext101_32x8d'¶

resnext50_32x4d = 'resnext50_32x4d'¶

shufflenet_v2_x0_5 = 'shufflenet_v2_x0_5'¶

shufflenet_v2_x1_0 = 'shufflenet_v2_x1_0'¶

shufflenet_v2_x1_5 = 'shufflenet_v2_x1_5'¶

shufflenet_v2_x2_0 = 'shufflenet_v2_x2_0'¶

squeezenet1_0 = 'squeezenet1_0'¶

squeezenet1_1 = 'squeezenet1_1'¶

vgg11 = 'vgg11'¶

vgg11_bn = 'vgg11_bn'¶

vgg13 = 'vgg13'¶

vgg13_bn = 'vgg13_bn'¶

vgg16 = 'vgg16'¶

vgg16_bn = 'vgg16_bn'¶

vgg19 = 'vgg19'¶

vgg19_bn = 'vgg19_bn'¶

wide_resnet101_2 = 'wide_resnet101_2'¶

wide_resnet50_2 = 'wide_resnet50_2'¶

SolverConfig¶

class rastervision.pytorch_learner.SolverConfig¶

Config related to solver aka optimizer.

lr¶

Learning rate. Defaults to 0.0001.

Type: PositiveFloat

num_epochs¶

Number of epochs (ie. sweeps through the whole training set). Defaults to 10.

Type: PositiveInt

test_num_epochs¶

Number of epochs to use in test mode. Defaults to 2.

Type: PositiveInt

test_batch_sz¶

Batch size to use in test mode. Defaults to 4.

Type: PositiveInt

overfit_num_steps¶

Number of optimizer steps to use in overfit mode. Defaults to 1.

Type: PositiveInt

sync_interval¶

The interval in epochs for each sync to the cloud. Defaults to 1.

Type: PositiveInt

batch_sz¶

Batch size. Defaults to 32.

Type: PositiveInt

one_cycle¶

If True, use triangular LR scheduler with a single cycle across all epochs with start and end LR being lr/10 and the peak being lr. Defaults to True.

Type: bool

multi_stage¶

List of epoch indices at which to divide LR by 10. Defaults to [].

Type: List

ClassificationDataFormat¶

class rastervision.pytorch_learner.ClassificationDataFormat¶

An enumeration.

image_folder = 'image_folder'¶

ClassificationDataConfig¶

class rastervision.pytorch_learner.ClassificationDataConfig¶

uri¶

URI of the dataset. This can be a zip file, a list of zip files, or a directory which contains a set of zip files. Defaults to None.

Type: Union[NoneType, str, List[str]]

train_sz¶

If set, the number of training images to use. If fewer images exist, then an exception will be raised. Defaults to None.

Type: Optional[int]

group_uris¶

This can be set instead of uri in order to specify groups of chips. Each element in the list is expected to be an object of the same form accepted by the uri field. The purpose of separating chips into groups is to be able to use the group_train_sz field. Defaults to None.

Type: Optional[List[Union[str, List[str]]]]

group_train_sz¶

If group_uris is set, this can be used to specify the number of chips to use per group. Defaults to None.

Type: Optional[int]

data_format¶

Defaults to <ClassificationDataFormat.image_folder: ‘image_folder’>.

Type: enum

class_names¶

Names of classes. Defaults to [].

Type: List[str]

class_colors¶

Colors used to display classes. Can be color 3-tuples in list form. Defaults to None.

Type: Union[NoneType, List[str], List[List]]

img_sz¶

Length of a side of each image in pixels. This is the size to transform it to during training, not the size in the raw dataset. Defaults to 256.

Type: PositiveInt

num_workers¶

Number of workers to use when DataLoader makes batches. Defaults to 4.

Type: int

augmentors¶

Names of albumentations augmentors to use for training batches. Choices include: [‘Blur’, ‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’, ‘GaussianBlur’, ‘GaussNoise’, ‘RGBShift’, ‘ToGray’]. Defaults to [‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’].

Type: List[str]

ClassificationModelConfig¶

class rastervision.pytorch_learner.ClassificationModelConfig¶

backbone¶

The torchvision.models backbone to use. Defaults to <Backbone.resnet18: ‘resnet18’>.

Type: enum

pretrained¶

If True, use ImageNet weights. If False, use random initialization. Defaults to True.

Type: bool

init_weights¶

URI of PyTorch model weights used to initialize model. If set, this supercedes the pretrained option. Defaults to None.

Type: Optional[str]

ClassificationLearnerConfig¶

class rastervision.pytorch_learner.ClassificationLearnerConfig¶

model¶

Type: ClassificationModelConfig

solver¶

Type: SolverConfig

data¶

Type: ClassificationDataConfig

predict_mode¶

If True, skips training, loads model, and does final eval. Defaults to False.

Type: bool

test_mode¶

If True, uses test_num_epochs, test_batch_sz, truncated datasets with only a single batch, image_sz that is cut in half, and num_workers = 0. This is useful for testing that code runs correctly on CPU without multithreading before running full job on GPU. Defaults to False.

Type: bool

overfit_mode¶

If True, uses half image size, and instead of doing epoch-based training, optimizes the model using a single batch repeatedly for overfit_num_steps number of steps. Defaults to False.

Type: bool

eval_train¶

If True, runs final evaluation on training set (in addition to test set). Useful for debugging. Defaults to False.

Type: bool

save_model_bundle¶

If True, saves a model bundle at the end of training which is zip file with model and this LearnerConfig which can be used to make predictions on new images at a later time. Defaults to True.

Type: bool

log_tensorboard¶

Save Tensorboard log files at the end of each epoch. Defaults to True.

Type: bool

run_tensorboard¶

run Tensorboard server during training. Defaults to False.

Type: bool

output_uri¶

URI of where to save output. Defaults to None.

Type: Optional[str]

SemanticSegmentationDataFormat¶

class rastervision.pytorch_learner.SemanticSegmentationDataFormat¶

An enumeration.

default = 'default'¶

SemanticSegmentationDataConfig¶

class rastervision.pytorch_learner.SemanticSegmentationDataConfig¶

uri¶

URI of the dataset. This can be a zip file, a list of zip files, or a directory which contains a set of zip files. Defaults to None.

Type: Union[NoneType, str, List[str]]

train_sz¶

If set, the number of training images to use. If fewer images exist, then an exception will be raised. Defaults to None.

Type: Optional[int]

group_uris¶

This can be set instead of uri in order to specify groups of chips. Each element in the list is expected to be an object of the same form accepted by the uri field. The purpose of separating chips into groups is to be able to use the group_train_sz field. Defaults to None.

Type: Optional[List[Union[str, List[str]]]]

group_train_sz¶

If group_uris is set, this can be used to specify the number of chips to use per group. Defaults to None.

Type: Optional[int]

data_format¶

Defaults to <SemanticSegmentationDataFormat.default: ‘default’>.

Type: enum

class_names¶

Names of classes. Defaults to [].

Type: List[str]

class_colors¶

Colors used to display classes. Can be color 3-tuples in list form. Defaults to None.

Type: Union[NoneType, List[str], List[List]]

img_sz¶

Length of a side of each image in pixels. This is the size to transform it to during training, not the size in the raw dataset. Defaults to 256.

Type: PositiveInt

num_workers¶

Number of workers to use when DataLoader makes batches. Defaults to 4.

Type: int

augmentors¶

Names of albumentations augmentors to use for training batches. Choices include: [‘Blur’, ‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’, ‘GaussianBlur’, ‘GaussNoise’, ‘RGBShift’, ‘ToGray’]. Defaults to [‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’].

Type: List[str]

SemanticSegmentationModelConfig¶

class rastervision.pytorch_learner.SemanticSegmentationModelConfig¶

backbone¶

The torchvision.models backbone to use. At the moment only resnet50 or resnet101 will work. Defaults to <Backbone.resnet50: ‘resnet50’>.

Type: enum

pretrained¶

If True, use ImageNet weights. If False, use random initialization. Defaults to True.

Type: bool

init_weights¶

URI of PyTorch model weights used to initialize model. If set, this supercedes the pretrained option. Defaults to None.

Type: Optional[str]

SemanticSegmentationLearnerConfig¶

class rastervision.pytorch_learner.SemanticSegmentationLearnerConfig¶

model¶

Type: SemanticSegmentationModelConfig

solver¶

Type: SolverConfig

data¶

Type: SemanticSegmentationDataConfig

predict_mode¶

If True, skips training, loads model, and does final eval. Defaults to False.

Type: bool

test_mode¶

If True, uses test_num_epochs, test_batch_sz, truncated datasets with only a single batch, image_sz that is cut in half, and num_workers = 0. This is useful for testing that code runs correctly on CPU without multithreading before running full job on GPU. Defaults to False.

Type: bool

overfit_mode¶

If True, uses half image size, and instead of doing epoch-based training, optimizes the model using a single batch repeatedly for overfit_num_steps number of steps. Defaults to False.

Type: bool

eval_train¶

If True, runs final evaluation on training set (in addition to test set). Useful for debugging. Defaults to False.

Type: bool

save_model_bundle¶

If True, saves a model bundle at the end of training which is zip file with model and this LearnerConfig which can be used to make predictions on new images at a later time. Defaults to True.

Type: bool

log_tensorboard¶

Save Tensorboard log files at the end of each epoch. Defaults to True.

Type: bool

run_tensorboard¶

run Tensorboard server during training. Defaults to False.

Type: bool

output_uri¶

URI of where to save output. Defaults to None.

Type: Optional[str]

ObjectDetectionDataFormat¶

class rastervision.pytorch_learner.ObjectDetectionDataFormat¶

An enumeration.

coco = 'coco'¶

ObjectDetectionDataConfig¶

class rastervision.pytorch_learner.ObjectDetectionDataConfig¶

uri¶

URI of the dataset. This can be a zip file, a list of zip files, or a directory which contains a set of zip files. Defaults to None.

Type: Union[NoneType, str, List[str]]

train_sz¶

If set, the number of training images to use. If fewer images exist, then an exception will be raised. Defaults to None.

Type: Optional[int]

group_uris¶

This can be set instead of uri in order to specify groups of chips. Each element in the list is expected to be an object of the same form accepted by the uri field. The purpose of separating chips into groups is to be able to use the group_train_sz field. Defaults to None.

Type: Optional[List[Union[str, List[str]]]]

group_train_sz¶

If group_uris is set, this can be used to specify the number of chips to use per group. Defaults to None.

Type: Optional[int]

data_format¶

Defaults to <ObjectDetectionDataFormat.coco: ‘coco’>.

Type: enum

class_names¶

Names of classes. Defaults to [].

Type: List[str]

class_colors¶

Colors used to display classes. Can be color 3-tuples in list form. Defaults to None.

Type: Union[NoneType, List[str], List[List]]

img_sz¶

Length of a side of each image in pixels. This is the size to transform it to during training, not the size in the raw dataset. Defaults to 256.

Type: PositiveInt

num_workers¶

Number of workers to use when DataLoader makes batches. Defaults to 4.

Type: int

augmentors¶

Names of albumentations augmentors to use for training batches. Choices include: [‘Blur’, ‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’, ‘GaussianBlur’, ‘GaussNoise’, ‘RGBShift’, ‘ToGray’]. Defaults to [‘RandomRotate90’, ‘HorizontalFlip’, ‘VerticalFlip’].

Type: List[str]

ObjectDetectionModelConfig¶

class rastervision.pytorch_learner.ObjectDetectionModelConfig¶

backbone¶

The torchvision.models backbone to use, which must be in the resnet* family. Defaults to <Backbone.resnet50: ‘resnet50’>.

Type: enum

pretrained¶

If True, use ImageNet weights. If False, use random initialization. Defaults to True.

Type: bool

init_weights¶

URI of PyTorch model weights used to initialize model. If set, this supercedes the pretrained option. Defaults to None.

Type: Optional[str]

ObjectDetectionLearnerConfig¶

class rastervision.pytorch_learner.ObjectDetectionLearnerConfig¶

model¶

Type: ObjectDetectionModelConfig

solver¶

Type: SolverConfig

data¶

Type: ObjectDetectionDataConfig

predict_mode¶

If True, skips training, loads model, and does final eval. Defaults to False.

Type: bool

test_mode¶

If True, uses test_num_epochs, test_batch_sz, truncated datasets with only a single batch, image_sz that is cut in half, and num_workers = 0. This is useful for testing that code runs correctly on CPU without multithreading before running full job on GPU. Defaults to False.

Type: bool

overfit_mode¶

If True, uses half image size, and instead of doing epoch-based training, optimizes the model using a single batch repeatedly for overfit_num_steps number of steps. Defaults to False.

Type: bool

eval_train¶

If True, runs final evaluation on training set (in addition to test set). Useful for debugging. Defaults to False.

Type: bool

save_model_bundle¶

If True, saves a model bundle at the end of training which is zip file with model and this LearnerConfig which can be used to make predictions on new images at a later time. Defaults to True.

Type: bool

log_tensorboard¶

Save Tensorboard log files at the end of each epoch. Defaults to True.

Type: bool

run_tensorboard¶

run Tensorboard server during training. Defaults to False.

Type: bool

output_uri¶

URI of where to save output. Defaults to None.

Type: Optional[str]