megadetector 5.0.12__py3-none-any.whl → 5.0.14__py3-none-any.whl

megadetector/postprocessing/compare_batch_results.py

@@ -43,29 +43,31 @@ class PairwiseBatchComparisonOptions:
     pairwise options sets is stored in the BatchComparisonsOptions class.
     """
     
-    #: First filename to compare
-    results_filename_a = None
-    
-    #: Second filename to compare
-    results_filename_b = None
-    
-    #: Description to use in the output HTML for filename A
-    results_description_a = None
-    
-    #: Description to use in the output HTML for filename B
-    results_description_b = None
-    
-    #: Per-class detection thresholds to use for filename A (including a 'default' threshold)
-    detection_thresholds_a = {'animal':0.15,'person':0.15,'vehicle':0.15,'default':0.15}
-    
-    #: Per-class detection thresholds to use for filename B (including a 'default' threshold)
-    detection_thresholds_b = {'animal':0.15,'person':0.15,'vehicle':0.15,'default':0.15}
-
-    #: Rendering threshold to use for all categories for filename A
-    rendering_confidence_threshold_a = 0.1
+    def __init__(self):
+        
+        #: First filename to compare
+        self.results_filename_a = None
+        
+        #: Second filename to compare
+        self.results_filename_b = None
+        
+        #: Description to use in the output HTML for filename A
+        self.results_description_a = None
+        
+        #: Description to use in the output HTML for filename B
+        self.results_description_b = None
+        
+        #: Per-class detection thresholds to use for filename A (including a 'default' threshold)
+        self.detection_thresholds_a = {'animal':0.15,'person':0.15,'vehicle':0.15,'default':0.15}
+        
+        #: Per-class detection thresholds to use for filename B (including a 'default' threshold)
+        self.detection_thresholds_b = {'animal':0.15,'person':0.15,'vehicle':0.15,'default':0.15}
     
-    #: Rendering threshold to use for all categories for filename B
-    rendering_confidence_threshold_b = 0.1
+        #: Rendering threshold to use for all categories for filename A
+        self.rendering_confidence_threshold_a = 0.1
+        
+        #: Rendering threshold to use for all categories for filename B
+        self.rendering_confidence_threshold_b = 0.1
 
 # ...class PairwiseBatchComparisonOptions
 
@@ -75,58 +77,60 @@ class BatchComparisonOptions:
     Defines the options for a set of (possibly many) pairwise comparisons.
     """
     
-    #: Folder to which we should write HTML output
-    output_folder = None
-    
-    #: Base folder for images (which are specified as relative files)
-    image_folder = None
-    
-    #: Job name to use in the HTML output file
-    job_name = ''
-    
-    #: Maximum number of images to render for each category, where a "category" here is
-    #: "detections_a_only", "detections_b_only", etc., or None to render all images.
-    max_images_per_category = 1000
-    
-    #: Maximum number of images per HTML page (paginates if a category page goes beyond this),
-    #: or None to disable pagination.
-    max_images_per_page = None
-    
-    #: Colormap to use for detections in file A (maps detection categories to colors)
-    colormap_a = ['Red']
-    
-    #: Colormap to use for detections in file B (maps detection categories to colors)
-    colormap_b = ['RoyalBlue']
-
-    #: Process-based parallelization isn't supported yet; this must be "True"
-    parallelize_rendering_with_threads = True
-    
-    #: List of filenames to include in the comparison, or None to use all files
-    filenames_to_include = None
-    
-    #: Compare only detections/non-detections, ignore categories (still renders categories)
-    class_agnostic_comparison = False
-    
-    #: Width of images to render in the output HTML
-    target_width = 800
-    
-    #: Number of workers to use for rendering, or <=1 to disable parallelization
-    n_rendering_workers = 20
-    
-    #: Random seed for image sampling (not used if max_images_per_category is None)
-    random_seed = 0
-    
-    #: Whether to sort results by confidence; if this is False, sorts by filename
-    sort_by_confidence = False
-    
-    #: The expectation is that all results sets being compared will refer to the same images; if this
-    #: is True (default), we'll error if that's not the case, otherwise non-matching lists will just be
-    #: a warning.
-    error_on_non_matching_lists = True
-    
-    #: List of PairwiseBatchComparisonOptions that defines the comparisons we'll render.
-    pairwise_options = []
+    def __init__(self):
+        
+        #: Folder to which we should write HTML output
+        self.output_folder = None
+        
+        #: Base folder for images (which are specified as relative files)
+        self.image_folder = None
+        
+        #: Job name to use in the HTML output file
+        self.job_name = ''
+        
+        #: Maximum number of images to render for each category, where a "category" here is
+        #: "detections_a_only", "detections_b_only", etc., or None to render all images.
+        self.max_images_per_category = 1000
+        
+        #: Maximum number of images per HTML page (paginates if a category page goes beyond this),
+        #: or None to disable pagination.
+        self.max_images_per_page = None
+        
+        #: Colormap to use for detections in file A (maps detection categories to colors)
+        self.colormap_a = ['Red']
+        
+        #: Colormap to use for detections in file B (maps detection categories to colors)
+        self.colormap_b = ['RoyalBlue']
     
+        #: Process-based parallelization isn't supported yet; this must be "True"
+        self.parallelize_rendering_with_threads = True
+        
+        #: List of filenames to include in the comparison, or None to use all files
+        self.filenames_to_include = None
+        
+        #: Compare only detections/non-detections, ignore categories (still renders categories)
+        self.class_agnostic_comparison = False
+        
+        #: Width of images to render in the output HTML
+        self.target_width = 800
+        
+        #: Number of workers to use for rendering, or <=1 to disable parallelization
+        self.n_rendering_workers = 20
+        
+        #: Random seed for image sampling (not used if max_images_per_category is None)
+        self.random_seed = 0
+        
+        #: Whether to sort results by confidence; if this is False, sorts by filename
+        self.sort_by_confidence = False
+        
+        #: The expectation is that all results sets being compared will refer to the same images; if this
+        #: is True (default), we'll error if that's not the case, otherwise non-matching lists will just be
+        #: a warning.
+        self.error_on_non_matching_lists = True
+        
+        #: List of PairwiseBatchComparisonOptions that defines the comparisons we'll render.
+        self.pairwise_options = []
+        
 # ...class BatchComparisonOptions
     
 
@@ -135,22 +139,24 @@ class PairwiseBatchComparisonResults:
     The results from a single pairwise comparison.
     """
     
-    #: String of HTML content suitable for rendering to an HTML file
-    html_content = None
-    
-    #: Possibly-modified version of the PairwiseBatchComparisonOptions supplied as input.
-    pairwise_options = None
+    def __init__(self):
     
-    #: A dictionary with keys including:
-    #:
-    #: common_detections
-    #: common_non_detections
-    #: detections_a_only
-    #: detections_b_only
-    #: class_transitions
-    #
-    #: Each of these maps a filename to a two-element list (the image in set A, the image in set B).
-    categories_to_image_pairs = None
+        #: String of HTML content suitable for rendering to an HTML file
+        self.html_content = None
+        
+        #: Possibly-modified version of the PairwiseBatchComparisonOptions supplied as input.
+        self.pairwise_options = None
+        
+        #: A dictionary with keys including:
+        #:
+        #: common_detections
+        #: common_non_detections
+        #: detections_a_only
+        #: detections_b_only
+        #: class_transitions
+        #
+        #: Each of these maps a filename to a two-element list (the image in set A, the image in set B).
+        self.categories_to_image_pairs = None
 
 # ...class PairwiseBatchComparisonResults
     
@@ -160,11 +166,13 @@ class BatchComparisonResults:
     The results from a set of pairwise comparisons
     """
     
-    #: Filename containing HTML output
-    html_output_file = None
-    
-    #: A list of PairwiseBatchComparisonResults
-    pairwise_results = None
+    def __init__(self):
+        
+        #: Filename containing HTML output
+        self.html_output_file = None
+        
+        #: A list of PairwiseBatchComparisonResults
+        self.pairwise_results = None
     
 # ...class BatchComparisonResults    
 

megadetector/postprocessing/convert_output_format.py

@@ -30,8 +30,11 @@ CONF_DIGITS = 3
 
 #%% Conversion functions
 
-def convert_json_to_csv(input_path,output_path=None,min_confidence=None,
-                        omit_bounding_boxes=False,output_encoding=None,
+def convert_json_to_csv(input_path,
+                        output_path=None,
+                        min_confidence=None,
+                        omit_bounding_boxes=False,
+                        output_encoding=None,
                         overwrite=True):
     """
     Converts a MD results .json file to a totally non-standard .csv format.
@@ -76,9 +79,9 @@ def convert_json_to_csv(input_path,output_path=None,min_confidence=None,
     # n_non_empty_detection_categories = len(annotation_constants.annotation_bbox_categories) - 1
     n_non_empty_detection_categories = annotation_constants.NUM_DETECTOR_CATEGORIES
     detection_category_column_names = []
-    assert annotation_constants.detector_bbox_categories[0] == 'empty'
+    assert annotation_constants.detector_bbox_category_id_to_name[0] == 'empty'
     for cat_id in range(1,n_non_empty_detection_categories+1):
-        cat_name = annotation_constants.detector_bbox_categories[cat_id]
+        cat_name = annotation_constants.detector_bbox_category_id_to_name[cat_id]
         detection_category_column_names.append('max_conf_' + cat_name)
     
     n_classification_categories = 0
@@ -370,6 +373,8 @@ def main():
     parser.add_argument('--output_path',type=str,default=None,
                         help='Output filename ending in .json or .csv (defaults to ' + \
                              'input file, with .json/.csv replaced by .csv/.json)')
+    parser.add_argument('--omit_bounding_boxes',action='store_true',
+                        help='Output bounding box text from .csv output (large and usually not useful)')        
     
     if len(sys.argv[1:]) == 0:
         parser.print_help()
@@ -386,9 +391,11 @@ def main():
             raise ValueError('Illegal input file extension')    
     
     if args.input_path.endswith('.csv') and args.output_path.endswith('.json'):
+        assert not args.omit_bounding_boxes, \
+            '--omit_bounding_boxes does not apply to csv --> json conversion'
         convert_csv_to_json(args.input_path,args.output_path)
     elif args.input_path.endswith('.json') and args.output_path.endswith('.csv'):
-        convert_json_to_csv(args.input_path,args.output_path)
+        convert_json_to_csv(args.input_path,args.output_path,omit_bounding_boxes=args.omit_bounding_boxes)
     else:
         raise ValueError('Illegal format combination')            
 

megadetector/postprocessing/merge_detections.py

@@ -31,26 +31,37 @@ class MergeDetectionsOptions:
     
     def __init__(self):
         
+        #: Maximum detection size to include in the merged output
         self.max_detection_size = 1.01
+        
+        #: Minimum detection size to include in the merged output
         self.min_detection_size = 0
+        
+        #: Exclude detections whose confidence in the source file(s) is less
+        #: than this.  Should have the same length as the number of source files.
         self.source_confidence_thresholds = [0.05]
         
-        # Don't bother merging into target images if there is a similar detection
-        # above this threshold (or if there is *any* detection above this threshold,
-        # and merge_empty_only is True)
+        #: Don't bother merging into target images if there is a similar detection
+        #: above this threshold (or if there is *any* detection above this threshold,
+        #: and merge_empty_only is True)
         self.target_confidence_threshold = 0.2
         
-        # If you want to merge only certain categories, specify one
-        # (but not both) of these.  These are category IDs, not names.
+        #: If you want to merge only certain categories, specify one
+        #: (but not both) of these.  These are category IDs, not names.
         self.categories_to_include = None
+        
+        #: If you want to merge only certain categories, specify one
+        #: (but not both) of these.  These are category IDs, not names.
         self.categories_to_exclude = None
 
-        # Only merge detections into images that have *no* detections in the 
-        # target results file.
+        #: Only merge detections into images that have *no* detections in the 
+        #: target results file.
         self.merge_empty_only = False
         
+        #: IoU threshold above which two detections are considered the same
         self.iou_threshold = 0.65
         
+        #: Error if this is False and the output file exists
         self.overwrite = False
 
 

megadetector/postprocessing/postprocess_batch_results.py

@@ -72,135 +72,139 @@ class PostProcessingOptions:
     Options used to parameterize process_batch_results().
     """
     
-    ### Required inputs
-
-    #: MD results .json file to process
-    md_results_file = ''
+    def __init__(self):
+        
+        ### Required inputs
     
-    #: Folder to which we should write HTML output
-    output_dir = ''
-
-    ### Options
-
-    #: Folder where images live (filenames in [md_results_file] should be relative to this folder)
-    image_base_dir = '.'
-
-    ## These apply only when we're doing ground-truth comparisons
+        #: MD results .json file to process
+        self.md_results_file = ''
+        
+        #: Folder to which we should write HTML output
+        self.output_dir = ''
     
-    #: Optional .json file containing ground truth information
-    ground_truth_json_file = ''
-
-    #: Classes we'll treat as negative
-    #:
-    #: Include the token "#NO_LABELS#" to indicate that an image with no annotations
-    #: should be considered empty.
-    negative_classes = DEFAULT_NEGATIVE_CLASSES
+        ### Options
     
-    #: Classes we'll treat as neither positive nor negative
-    unlabeled_classes = DEFAULT_UNKNOWN_CLASSES
-
-    #: A list of output sets that we should count, but not render images for.
-    #:
-    #: Typically used to preview sets with lots of empties, where you don't want to
-    #: subset but also don't want to render 100,000 empty images.
-    #:
-    #: detections, non_detections
-    #: detections_animal, detections_person, detections_vehicle
-    rendering_bypass_sets = []
-
-    #: If this is None, choose a confidence threshold based on the detector version.
-    #:
-    #: This can either be a float or a dictionary mapping category names (not IDs) to 
-    #: thresholds.  The category "default" can be used to specify thresholds for 
-    #: other categories.  Currently the use of a dict here is not supported when 
-    #: ground truth is supplied.
-    confidence_threshold = None
+        #: Folder where images live (filenames in [md_results_file] should be relative to this folder)
+        self.image_base_dir = '.'
     
-    #: Confidence threshold to apply to classification (not detection) results
-    #:
-    #: Only a float is supported here (unlike the "confidence_threshold" parameter, which
-    #: can be a dict).    
-    classification_confidence_threshold = 0.5
-
-    #: Used for summary statistics only
-    target_recall = 0.9
-
-    #: Number of images to sample, -1 for "all images"
-    num_images_to_sample = 500
-
-    #: Random seed for sampling, or None
-    sample_seed = 0 # None
-
-    #: Image width for images in the HTML output
-    viz_target_width = 800
-
-    #: Line width (in pixels) for rendering detections
-    line_thickness = 4
+        ## These apply only when we're doing ground-truth comparisons
+        
+        #: Optional .json file containing ground truth information
+        self.ground_truth_json_file = ''
+    
+        #: Classes we'll treat as negative
+        #:
+        #: Include the token "#NO_LABELS#" to indicate that an image with no annotations
+        #: should be considered empty.
+        self.negative_classes = DEFAULT_NEGATIVE_CLASSES
+        
+        #: Classes we'll treat as neither positive nor negative
+        self.unlabeled_classes = DEFAULT_UNKNOWN_CLASSES
+    
+        #: A list of output sets that we should count, but not render images for.
+        #:
+        #: Typically used to preview sets with lots of empties, where you don't want to
+        #: subset but also don't want to render 100,000 empty images.
+        #:
+        #: detections, non_detections
+        #: detections_animal, detections_person, detections_vehicle
+        self.rendering_bypass_sets = []
+    
+        #: If this is None, choose a confidence threshold based on the detector version.
+        #:
+        #: This can either be a float or a dictionary mapping category names (not IDs) to 
+        #: thresholds.  The category "default" can be used to specify thresholds for 
+        #: other categories.  Currently the use of a dict here is not supported when 
+        #: ground truth is supplied.
+        self.confidence_threshold = None
+        
+        #: Confidence threshold to apply to classification (not detection) results
+        #:
+        #: Only a float is supported here (unlike the "confidence_threshold" parameter, which
+        #: can be a dict).    
+        self.classification_confidence_threshold = 0.5
     
-    #: Box expansion (in pixels) for rendering detections
-    box_expansion = 0
-
-    #: Job name to include in big letters in the output HTML
-    job_name_string = None
+        #: Used for summary statistics only
+        self.target_recall = 0.9
     
-    #: Model version string to include in the output HTML
-    model_version_string = None
+        #: Number of images to sample, -1 for "all images"
+        self.num_images_to_sample = 500
     
-    #: Sort order for the output, should be one of "filename", "confidence", or "random"
-    html_sort_order = 'filename'
-   
-    #: If True, images in the output HTML will be links back to the original images
-    link_images_to_originals = True
+        #: Random seed for sampling, or None
+        self.sample_seed = 0 # None
     
-    #: Optionally separate detections into categories (animal/vehicle/human)
-    #: 
-    #: Currently only supported when ground truth is unavailable
-    separate_detections_by_category = True
-
-    #: Optionally replace one or more strings in filenames with other strings;
-    #: useful for taking a set of results generated for one folder structure
-    #: and applying them to a slightly different folder structure.
-    api_output_filename_replacements = {}
+        #: Image width for images in the HTML output
+        self.viz_target_width = 800
     
-    #: Optionally replace one or more strings in filenames with other strings;
-    #: useful for taking a set of results generated for one folder structure
-    #: and applying them to a slightly different folder structure.
-    ground_truth_filename_replacements = {}
-
-    #: Allow bypassing API output loading when operating on previously-loaded
-    #: results.  If present, this is a Pandas DataFrame.  Almost never useful.
-    api_detection_results = None
+        #: Line width (in pixels) for rendering detections
+        self.line_thickness = 4
+        
+        #: Box expansion (in pixels) for rendering detections
+        self.box_expansion = 0
     
-    #: Allow bypassing API output loading when operating on previously-loaded
-    #: results.  If present, this is a str --> obj dict.  Almost never useful.
-    api_other_fields = None
-
-    #: Should we also split out a separate report about the detections that were
-    #: just below our main confidence threshold?
-    #:
-    #: Currently only supported when ground truth is unavailable.
-    include_almost_detections = False
+        #: Job name to include in big letters in the output HTML
+        self.job_name_string = None
+        
+        #: Model version string to include in the output HTML
+        self.model_version_string = None
+        
+        #: Sort order for the output, should be one of "filename", "confidence", or "random"
+        self.html_sort_order = 'filename'
+       
+        #: If True, images in the output HTML will be links back to the original images
+        self.link_images_to_originals = True
+        
+        #: Optionally separate detections into categories (animal/vehicle/human)
+        #: 
+        #: Currently only supported when ground truth is unavailable
+        self.separate_detections_by_category = True
+    
+        #: Optionally replace one or more strings in filenames with other strings;
+        #: useful for taking a set of results generated for one folder structure
+        #: and applying them to a slightly different folder structure.
+        self.api_output_filename_replacements = {}
+        
+        #: Optionally replace one or more strings in filenames with other strings;
+        #: useful for taking a set of results generated for one folder structure
+        #: and applying them to a slightly different folder structure.
+        self.ground_truth_filename_replacements = {}
+    
+        #: Allow bypassing API output loading when operating on previously-loaded
+        #: results.  If present, this is a Pandas DataFrame.  Almost never useful.
+        self.api_detection_results = None
+        
+        #: Allow bypassing API output loading when operating on previously-loaded
+        #: results.  If present, this is a str --> obj dict.  Almost never useful.
+        self.api_other_fields = None
+    
+        #: Should we also split out a separate report about the detections that were
+        #: just below our main confidence threshold?
+        #:
+        #: Currently only supported when ground truth is unavailable.
+        self.include_almost_detections = False
+        
+        #: Only a float is supported here (unlike the "confidence_threshold" parameter, which
+        #: can be a dict).
+        self.almost_detection_confidence_threshold = None
     
-    #: Only a float is supported here (unlike the "confidence_threshold" parameter, which
-    #: can be a dict).
-    almost_detection_confidence_threshold = None
+        #: Enable/disable rendering parallelization    
+        self.parallelize_rendering = False
+        
+        #: Number of threads/processes to use for rendering parallelization
+        self.parallelize_rendering_n_cores = 25
+        
+        #: Whether to use threads (True) or processes (False) for rendering parallelization
+        self.parallelize_rendering_with_threads = True
+        
+        #: When classification results are present, should be sort alphabetically by class name (False)
+        #: or in descending order by frequency (True)?
+        self.sort_classification_results_by_count = False    
+        
+        #: Should we split individual pages up into smaller pages if there are more than
+        #: N images?
+        self.max_figures_per_html_file = None
 
-    #: Enable/disable rendering parallelization    
-    parallelize_rendering = False
-    
-    #: Number of threads/processes to use for rendering parallelization
-    parallelize_rendering_n_cores = 25
-    
-    #: Whether to use threads (True) or processes (False) for rendering parallelization
-    parallelize_rendering_with_threads = True
-    
-    #: When classification results are present, should be sort alphabetically by class name (False)
-    #: or in descending order by frequency (True)?
-    sort_classification_results_by_count = False    
-    
-    #: Should we split individual pages up into smaller pages if there are more than
-    #: N images?
-    max_figures_per_html_file = None
+    # ...__init__()
     
 # ...PostProcessingOptions
 
@@ -210,14 +214,16 @@ class PostProcessingResults:
     Return format from process_batch_results
     """
     
-    #: HTML file to which preview information was written
-    output_html_file = ''
-    
-    #: Pandas Dataframe containing detection results
-    api_detection_results = None
-    
-    #: str --> obj dictionary containing other information loaded from the results file
-    api_other_fields = None
+    def __init__(self):
+        
+        #: HTML file to which preview information was written
+        self.output_html_file = ''
+        
+        #: Pandas Dataframe containing detection results
+        self.api_detection_results = None
+        
+        #: str --> obj dictionary containing other information loaded from the results file
+        self.api_other_fields = None
 
 
 ##%% Helper classes and functions
@@ -901,8 +907,8 @@ def process_batch_results(options):
             this set of results; see the PostProcessingOptions class for details.
            
     Returns:
-        PostProcessingResults: information about the results/preview, most importantly the HTML filename
-            of the output.  See the PostProcessingResults class for details.
+        PostProcessingResults: information about the results/preview, most importantly the 
+        HTML filename of the output.  See the PostProcessingResults class for details.
     """
     ppresults = PostProcessingResults()