def __init__(self, json_result):
from trustedanalytics import get_frame
self.num_features = json_result['num_features']
self.num_classes = json_result['num_classes']
self.covariance_matrix = None
if pandas_available:
self.summary_table = pd.DataFrame(data=json_result['coefficients'].values(),
columns=['coefficients'],
index=json_result['coefficients'].keys())
self.summary_table['degrees_freedom'] = pd.Series(data=json_result['degrees_freedom'].values(),
index=json_result['degrees_freedom'].keys())
if json_result.get('covariance_matrix', None) is not None:
self.summary_table['standard_errors'] = pd.Series(data=json_result['standard_errors'].values(),
index=json_result['standard_errors'].keys())
self.summary_table['wald_statistic'] = pd.Series(data=json_result['wald_statistic'].values(),
index=json_result['wald_statistic'].keys())
self.summary_table['p_value'] = pd.Series(data=json_result['p_value'].values(),
index=json_result['p_value'].keys())
self.covariance_matrix = get_frame(json_result['covariance_matrix']['uri'])
else:
self.summary_table = {
'coefficients': json_result['coefficients'],
'degrees_freedom': json_result['degrees_freedom']
}
if json_result.get('covariance_matrix', None) is not None:
self.summary_table['standard_errors'] = json_result['standard_errors']
self.summary_table['wald_statistic'] = json_result['wald_statistic']
self.summary_table['p_value'] = json_result['p_value']
self.summary_table['covariance_matrix'] = get_frame(json_result['covariance_matrix']['uri'])
def return_page_rank(json_result):
from trustedanalytics import get_frame
vertex_json = json_result["vertex_dictionary_output"]
edge_json = json_result["edge_dictionary_output"]
vertex_dictionary = dict([(k,get_frame(v["uri"])) for k,v in vertex_json.items()])
edge_dictionary = dict([(k,get_frame(v["uri"])) for k,v in edge_json.items()])
return {'vertex_dictionary': vertex_dictionary, 'edge_dictionary': edge_dictionary}
def return_page_rank(json_result):
from trustedanalytics import get_frame
vertex_json = json_result["vertex_dictionary_output"]
edge_json = json_result["edge_dictionary_output"]
vertex_dictionary = dict([(k,get_frame(v["uri"])) for k,v in vertex_json.items()])
edge_dictionary = dict([(k,get_frame(v["uri"])) for k,v in edge_json.items()])
return {'vertex_dictionary': vertex_dictionary, 'edge_dictionary': edge_dictionary}
def return_collaborative_filtering(json_result):
from trustedanalytics import get_frame
user_frame = get_frame(json_result['user_frame']['uri'])
item_frame = get_frame(json_result['item_frame']['uri'])
return {
'user_frame': user_frame,
'item_frame': item_frame,
'report': json_result['report']
}
def return_lda_train(selfish, json_result):
from trustedanalytics import get_frame
doc_frame = get_frame(json_result['doc_results']['id'])
word_frame = get_frame(json_result['word_results']['id'])
return {
'doc_results': doc_frame,
'word_results': word_frame,
'report': json_result['report']
}
def return_lda_train(json_result):
from trustedanalytics import get_frame
doc_frame = get_frame(json_result['topics_given_doc']['uri'])
word_frame = get_frame(json_result['word_given_topics']['uri'])
topic_frame = get_frame(json_result['topics_given_word']['uri'])
return {
'topics_given_doc': doc_frame,
'word_given_topics': word_frame,
'topics_given_word': topic_frame,
'report': json_result['report']
}
def return_clustering_coefficient(json_result):
from trustedanalytics import get_frame
from trustedanalytics.core.clusteringcoefficient import ClusteringCoefficient
if json_result.has_key('frame'):
frame = get_frame(json_result['frame']['uri'])
else:
frame = None
return ClusteringCoefficient(json_result['global_clustering_coefficient'], frame)
def return_clustering_coefficient(json_result):
from trustedanalytics import get_frame
from trustedanalytics.core.clusteringcoefficient import ClusteringCoefficient
if json_result.has_key('frame'):
frame = get_frame(json_result['frame']['uri'])
else:
frame = None
return ClusteringCoefficient(json_result['global_clustering_coefficient'], frame)
def return_principal_components_predict(selfish, json_result):
from trustedanalytics import get_frame
train_output = {
'output_frame': get_frame(json_result['output_frame']['id'])
}
if json_result.get('t_squared_index', None) is not None:
train_output['t_squared_index'] = json_result['t_squared_index']
return train_output
def return_loopy_belief_propagation(json_result):
from trustedanalytics import get_frame
vertex_json = json_result['frame_dictionary_output']
vertex_dictionary = dict([(k, get_frame(v["uri"]))
for k, v in vertex_json.items()])
return {
'vertex_dictionary': vertex_dictionary,
'time': json_result['time']
}
def return_power_iteration_clustering_predict(json_result):
from trustedanalytics import get_frame
predicted_frame = get_frame(json_result['frame']['uri'])
number_of_clusters = json_result['k']
cluster_size = json_result['cluster_size']
return {
'predicted_frame': predicted_frame,
'number_of_clusters': number_of_clusters,
'cluster_size': cluster_size
}
def get_frame(name):
global frame
if mode is None or mode == 'local':
print('Warning: Not connected to ATK')
return
if not frame is None:
return frame
frames = tap.get_frame_names()
if name in frames:
return tap.get_frame(name)
frame = tap.Frame(tap.UploadRows([], schema))
frame.name = name
return frame
def return_loopy_belief_propagation(json_result):
from trustedanalytics import get_frame
vertex_json = json_result['frame_dictionary_output']
vertex_dictionary = dict([(k,get_frame(v["uri"])) for k,v in vertex_json.items()])
return {'vertex_dictionary': vertex_dictionary, 'time': json_result['time']}
def return_connected_components(json_result):
from trustedanalytics import get_frame
dictionary = json_result["frame_dictionary_output"]
return dict([(k,get_frame(v["uri"])) for k,v in dictionary.items()])
def return_collaborative_filtering(json_result):
from trustedanalytics import get_frame
user_frame = get_frame(json_result['user_frame']['uri'])
item_frame= get_frame(json_result['item_frame']['uri'])
return { 'user_frame': user_frame, 'item_frame': item_frame, 'report': json_result['report'] }
def return_label_propagation(json_result):
from trustedanalytics import get_frame
frame = get_frame(json_result['output_frame']['uri'])
return { 'frame': frame, 'report': json_result['report'] }
def return_power_iteration_clustering_predict(json_result):
from trustedanalytics import get_frame
predicted_frame = get_frame(json_result['frame']['uri'])
number_of_clusters = json_result['k']
cluster_size = json_result['cluster_size']
return {'predicted_frame': predicted_frame , 'number_of_clusters': number_of_clusters, 'cluster_size': cluster_size}
def return_connected_components(json_result):
from trustedanalytics import get_frame
dictionary = json_result["frame_dictionary_output"]
return dict([(k,get_frame(v["uri"])) for k,v in dictionary.items()])
def return_loopy_belief_propagation(json_result):
from trustedanalytics import get_frame
frame = get_frame(json_result['output_frame']['uri'])
return { 'frame': frame, 'report': json_result['report'] }
def return_label_propagation(selfish, json_result):
from trustedanalytics import get_frame
frame = get_frame(json_result['output_frame']['id'])
return {'frame': frame, 'report': json_result['report']}
def _get_frame(uri_dict):
from trustedanalytics import get_frame
return get_frame(uri_dict['uri'])
def _get_frame(uri_dict):
from trustedanalytics import get_frame
return get_frame(uri_dict['uri'])
def return_giraph_lda_train(json_result):
from trustedanalytics import get_frame
doc_frame = get_frame(json_result['topics_given_doc']['uri'])
word_frame= get_frame(json_result['word_given_topics']['uri'])
topic_frame= get_frame(json_result['topics_given_word']['uri'])
return { 'topics_given_doc': doc_frame, 'word_given_topics': word_frame, 'topics_given_word': topic_frame, 'report': json_result['report'] }
def return_principal_components_predict(json_result):
from trustedanalytics import get_frame
train_output = {'output_frame': get_frame(json_result['output_frame']['uri']) }
if json_result.get('t_squared_index', None) is not None:
train_output['t_squared_index'] = json_result['t_squared_index']
return train_output
