def get_anchor(request, pk):
job = Job.objects.filter(pk=pk)[0]
exp, _ = Explanation.objects.get_or_create(type=ExplanationTypes.ANCHOR.value, split=job.split,
predictive_model=job.predictive_model, job=job)
exp.save()
result = explanation(exp.id)
return Response(result, status=200)
def get_shap(request, pk, explanation_target, prefix_target):
job = Job.objects.filter(pk=pk)[0]
exp, _ = Explanation.objects.get_or_create(
type=ExplanationTypes.SHAP.value,
split=job.split,
predictive_model=job.predictive_model,
job=job)
exp.save()
if 'shap' not in exp.results:
exp.results.update({'shap': dict()})
if explanation_target not in exp.results['shap']:
exp.results['shap'] = {explanation_target: dict()}
if explanation_target in exp.results[
'shap'] and prefix_target in exp.results['shap'][
explanation_target].keys():
return Response(pd.read_json(
exp.results['shap'][explanation_target][prefix_target],
typ='series',
orient='records'),
status=200)
else:
result = explanation(exp.id, explanation_target, prefix_target)
exp.results['shap'][explanation_target].update(
{prefix_target: pd.Series(result).to_json(orient='values')})
exp.save()
return Response(result, status=200)
def get_lime(request, pk, explanation_target):
job = Job.objects.filter(pk=pk)[0]
exp, _ = Explanation.objects.get_or_create(
type=ExplanationTypes.LIME.value,
split=job.split,
predictive_model=job.predictive_model,
job=job)
exp.save()
if 'lime' not in exp.results:
exp.results.update({'lime': dict()})
if explanation_target in exp.results['lime']:
return Response(exp.results['lime'][explanation_target], status=200)
else:
error, result = explanation(exp.id, explanation_target)
exp.results['lime'].update({explanation_target: result})
exp.save()
if error == 'True':
return Response(
{
'error':
'Explanation Target cannot be greater than ' + str(result)
},
status=status.HTTP_416_REQUESTED_RANGE_NOT_SATISFIABLE)
else:
return Response(result, status=200)
def get_cmfeedback(request, pk, top_k):
job = Job.objects.filter(pk=pk)[0]
exp, _ = Explanation.objects.get_or_create(
type=ExplanationTypes.CMFEEDBACK.value,
split=job.split,
predictive_model=job.predictive_model,
job=job)
exp.save()
result = explanation(exp.id, int(top_k))
return Response(result, status=200)
def get_shap(request, pk, explanation_target):
job = Job.objects.filter(pk=pk)[0]
exp, _ = Explanation.objects.get_or_create(
type=ExplanationTypes.SHAP.value,
split=job.split,
predictive_model=job.predictive_model,
job=job)
exp.save()
result = explanation(exp.id, explanation_target)
return Response(result, status=200)
def get_retrain(request, pk):
job = Job.objects.filter(pk=pk)[0]
exp, _ = Explanation.objects.get_or_create(
type=ExplanationTypes.RETRAIN.value,
split=job.split,
predictive_model=job.predictive_model,
job=job)
exp.save()
target = request.data
result = explanation(exp.id, target)
return Response(result, status=200)
def get_anchor(request, pk):
job = Job.objects.filter(pk=pk)[0]
exp, _ = Explanation.objects.get_or_create(
type=ExplanationTypes.ANCHOR.value,
split=job.split,
predictive_model=job.predictive_model,
job=job)
exp.save()
if 'anchor' in exp.results:
return Response(exp.results['anchor'], status=200)
else:
result = explanation(exp.id, explanation_target=None)
exp.results['anchor'] = result
exp.save()
return Response(result, status=200)
def get_ice(request, pk, explanation_target):
job = Job.objects.filter(pk=pk)[0]
exp, _ = Explanation.objects.get_or_create(
type=ExplanationTypes.ICE.value,
split=job.split,
predictive_model=job.predictive_model,
job=job)
exp.save()
if 'ice' not in exp.results:
exp.results.update({'ice': dict()})
if explanation_target in exp.results['ice']:
return Response(exp.results['ice'][explanation_target], status=200)
else:
result = explanation(exp.id, explanation_target)
exp.results['ice'].update({explanation_target: result})
exp.save()
return Response(result, status=200)
def progetto_padova():
JOB = Job.objects.get_or_create(
status=JobStatuses.CREATED.value,
type=JobTypes.PREDICTION.value,
split=Split.objects.get_or_create( # this creates the split of the log
type=SplitTypes.SPLIT_DOUBLE.value,
train_log=create_log( # this imports the log
import_log(BASE_DIR + RELATIVE_TRAIN_PATH),
RELATIVE_TRAIN_PATH,
BASE_DIR,
import_in_cache=False),
test_log=create_log( # this imports the log
import_log(BASE_DIR + RELATIVE_VALIDATION_PATH),
RELATIVE_VALIDATION_PATH,
BASE_DIR,
import_in_cache=False))[0],
encoding=Encoding.objects.
get_or_create( # this defines the encoding method
data_encoding=DataEncodings.LABEL_ENCODER.value,
value_encoding=ValueEncodings.SIMPLE_INDEX.value,
add_elapsed_time=False,
add_remaining_time=False,
add_executed_events=False,
add_resources_used=False,
add_new_traces=False,
prefix_length=5,
padding=True,
task_generation_type=TaskGenerationTypes.ALL_IN_ONE.value,
features=[])[0],
labelling=Labelling.objects.get_or_create( # this defines the label
type=LabelTypes.ATTRIBUTE_STRING.value,
attribute_name='label',
threshold_type=None,
threshold=None)[0],
clustering=Clustering.init(ClusteringMethods.NO_CLUSTER.value,
configuration={}),
predictive_model=PredictiveModel.
init( # this defines the predictive model
get_prediction_method_config(
PredictiveModels.CLASSIFICATION.value,
ClassificationMethods.DECISION_TREE.value,
payload={
'max_depth': 2,
'min_samples_split': 2,
'min_samples_leaf': 2
})),
hyperparameter_optimizer=HyperparameterOptimization.init(
{ # this defines the hyperparameter optimisation procedure
'type': HyperparameterOptimizationMethods.HYPEROPT.value,
'max_evaluations': 10,
'performance_metric': HyperOptAlgorithms.TPE.value,
'algorithm_type': HyperOptLosses.AUC.value
}),
create_models=True)[0]
# load log
train_log, test_log, additional_columns = get_train_test_log(JOB.split)
# encode
train_df, test_df = encode_label_logs(train_log, test_log, JOB)
# train + evaluate
results, model_split = MODEL[JOB.predictive_model.predictive_model][
ModelActions.BUILD_MODEL_AND_TEST.value](train_df, test_df,
_init_clusterer(
JOB.clustering, train_df),
JOB)
if JOB.create_models:
save_models(model_split, JOB)
# predict
data_df = pd.concat([train_df, test_df])
results = MODEL[JOB.predictive_model.predictive_model][
ModelActions.PREDICT.value](JOB, data_df)
results = MODEL[JOB.predictive_model.predictive_model][
ModelActions.PREDICT_PROBA.value](JOB, data_df)
# lime
exp = Explanation.objects.get_or_create(
type=ExplanationTypes.LIME.value,
split=JOB.
split, # this defines the analysed log, you can use a different one from the training one
predictive_model=JOB.predictive_model,
job=JOB)[0]
error, result = explanation(exp.id, int(EXPLANATION_TARGET))
