Expand Up @@ -380,15 +380,15 @@ def analyze_videos( trainingsetindex=trainingsetindex, overwrite=False, identity_only=identity_only, destfolder=str(destfolder), destfolder=str(output_path), ) stitch_tracklets( config, [str(video)], videotype, shuffle, trainingsetindex, destfolder=str(destfolder), destfolder=str(output_path), )
print( Expand Down Expand Up @@ -436,14 +436,25 @@ def create_df_from_prediction( ) if pred_unique_bodyparts is not None: coordinate_labels_unique = ["x", "y", "likelihood"] results_unique_df_index = pd.MultiIndex.from_product( [ [dlc_scorer], auxiliaryfunctions.get_unique_bodyparts(cfg), coordinate_labels_unique, ], names=["scorer", "bodyparts", "coords"], ) if n_individuals > 1: results_unique_df_index = pd.MultiIndex.from_product( [ [dlc_scorer], ['single'], auxiliaryfunctions.get_unique_bodyparts(cfg), coordinate_labels_unique, ], names=["scorer", "individuals", "bodyparts", "coords"], ) else: results_unique_df_index = pd.MultiIndex.from_product( [ [dlc_scorer], auxiliaryfunctions.get_unique_bodyparts(cfg), coordinate_labels_unique, ], names=["scorer", "bodyparts", "coords"], ) df_u = pd.DataFrame( pred_unique_bodyparts.reshape((len(pred_unique_bodyparts), -1)), columns=results_unique_df_index, Expand Down