identify.predict_ids

identify.predict_ids(reference_dict, query_dict, id_df, proposed_id_count=10)

Predict the identities of individuals in the query set

Return a DataFrame of the most proposed_id_count most similar individuals in the reference set for each query image, along with their cosine similarity.

Parameters

Name	Type	Description	Default
reference_dict	dict	Dictionary where the key is the reference image’s name. The value associate with each key is a NumPy array of shape (M, ) where M is the number of features in the feature vector.	required
query_dict	dict	Dictionary where the key is the query image’s name. The value associate with each key is a NumPy array of shape (M, ) where M is the number of features in the feature vector.	required
id_df	pd.DataFrame	DataFrame containing the identities, individual_id, and image file name, image, for every image in the reference set.	required
proposed_id_count	integer	The number of proposed IDs to return for each query image.	10

Examples

>>> import numpy as np
>>> import pandas as pd
>>> from pyseter.identify import predict_ids
>>> 
>>> ref_dict = {'image1': np.array([0.1, 0.11])}
>>> query_dict = {'image2': np.array([0.1, 0.12])}
>>> id_df = pd.DataFrame({'image': 'image1', 'individual_id': 'a'})
>>> 
>>> results = predict_ids(ref_dict, query_dict, id_df, proposed_id_count=1)
>>> len(results)
1