delta.config.Config

class delta.config.Config(presets, models, model_file_rois, model_file_seg, model_file_track, target_size_rois, target_size_seg, target_size_track, training_set_rois, training_set_seg, training_set_track, eval_movie, rotation_correction, drift_correction, whole_frame_drift, crop_windows, min_roi_area, min_cell_area, memory_growth_limit, pipeline_seg_batch, pipeline_track_batch, pipeline_chunk_size, number_of_cores, tolerable_resizing_rois)

Configuration class containing DeLTA parameters.

__init__(presets, models, model_file_rois, model_file_seg, model_file_track, target_size_rois, target_size_seg, target_size_track, training_set_rois, training_set_seg, training_set_track, eval_movie, rotation_correction, drift_correction, whole_frame_drift, crop_windows, min_roi_area, min_cell_area, memory_growth_limit, pipeline_seg_batch, pipeline_track_batch, pipeline_chunk_size, number_of_cores, tolerable_resizing_rois)

Methods

__init__(presets, models, model_file_rois, ...)
apply_tensorflow_config()	Apply the tensorflow-relevant parts of the config object.
compare(other[, level])	Return or print the list of differences between two Config objects.
default(presets)	Return the default config for this preset.
demo_movie_path()	Return the path of the demo movie for the specified presets.
model(model)	Return the keras model for the specified preset and model type.
read(path)	Read a configuration file.
sanitize()	Return a serializable version of the config.
training_set_path(model)	Return the path of the training set for the specified model.
update(key, value)	Update in place the configuration with the given key and value.
write(path)	Write a configuration file.

Attributes

presets	Type of analysis: can be '2D' or 'mothermachine'.
models	Which models need to be loaded ('rois', 'seg' or 'track').
model_file_rois	Path to the model file for ROI segmentation.
model_file_seg	Path to the model file for cell segmentation.
model_file_track	Path to the model file for cell tracking.
target_size_rois	Input size of the ROI segmentation model.
target_size_seg	Input size of the cell segmentation model.
target_size_track	Input size of the cell tracking model.
training_set_rois	Path to the ROI segmentation training set.
training_set_seg	Path to the cell segmentation training set.
training_set_track	Path to the cell tracking training set.
eval_movie	Path to the evaluation movie.
rotation_correction	Whether or not to correct for image rotation (mothermachine only).
drift_correction	Whether or not to correct for drift over time (mothermachine only).
whole_frame_drift	If correcting for drift, use the entire frame instead of the region above the chambers.
crop_windows	If True, crop input images into windows of size target_size_seg for segmentation, otherwise resize them.
min_roi_area	Minimum area of detected ROIs in pixels (mothermachine only).
min_cell_area	Minimum area of detected cells in pixels.
memory_growth_limit	If running into OOM issues or having troupble with CuDNN loading, try setting this to a value in MB: 1024, 2048, etc.
pipeline_seg_batch	If running into OOM issues during segmentation with the pipeline, try lowering this value.
pipeline_track_batch	If running into OOM issues during segmentation with the pipeline, try lowering this value.
pipeline_chunk_size	If running into OOM issues during segmentation with the pipeline, try lowering this value.
number_of_cores	This will limit the number of cores TensorFlow can use.
tolerable_resizing_rois	Maximum tolerable resizing factor of the template for ROI identification.