print("test", test_score)
print(classification_report(y_test, pipe.predict(X_test)))
data[labels.name + ' probability'] = pipe.predict_proba(X)[:, 1]
# Create LIME Explainer
feature_names = list(ce.columns_)
categorical_features = list(ce.cat_columns_ix_.values())
categorical_names = {i: ce.classes_[c]
for c, i in ce.cat_columns_ix_.items()}
class_names = ['No ' + labels.name, labels.name]
explainer = LimeTabularExplainer(ce.transform(data),
feature_names=feature_names,
class_names=class_names,
categorical_features=categorical_features,
categorical_names=categorical_names)
# Create and save the combined Logistic Regression and LIME Explained Model.
explainedmodel = ExplainedModel(data=data, labels=labels, model_name='telco_linear',
categoricalencoder=ce, pipeline=pipe,
explainer=explainer,data_dir=data_dir)
explainedmodel.save()
# If running as as experiment, this will track the metrics and add the model trained in this training run to the experiment history.
cdsw.track_metric("train_score",round(train_score,2))
cdsw.track_metric("test_score",round(test_score,2))
cdsw.track_metric("model_path",explainedmodel.model_path)
cdsw.track_file(explainedmodel.model_path)
#import os, sys
from collections import ChainMap
#from pandas.io.json import dumps as jsonify
from churnexplainer import ExplainedModel
em = ExplainedModel(model_name='telco_linear',data_dir='/home/cdsw')
def explain(args):
#data = dict(ChainMap(request.args, em.default_data))
data = dict(ChainMap(args, em.default_data))
data = em.cast_dct(data)
probability, explanation = em.explain_dct(data)
return {
'data': dict(data),
'probability': probability,
'explanation': explanation
}
#test
#x={"StreamingTV":"No","MonthlyCharges":70.35,"PhoneService":"No","PaperlessBilling":"No","Partner":"No","OnlineBackup":"No","gender":"Female","Contract":"Month-to-month","TotalCharges":1397.475,"StreamingMovies":"No","DeviceProtection":"No","PaymentMethod":"Bank transfer (automatic)","tenure":29,"Dependents":"No","OnlineSecurity":"No","MultipleLines":"No","InternetService":"DSL","SeniorCitizen":"No","TechSupport":"No"}
#explain(x)
log = logging.getLogger("werkzeug")
log.setLevel(logging.ERROR)
# Since we have access in an environment variable, we want to write it to our UI
# Change the line in the flask/single_view.html file.
if os.environ.get("SHTM_ACCESS_KEY") != None:
access_key = os.environ.get("SHTM_ACCESS_KEY", "")
subprocess.call([
"sed",
"-i",
's/const\saccessKey.*/const accessKey = "' + access_key + '";/',
"/home/cdsw/flask/single_view.html",
])
# Load the explained model
em = ExplainedModel(model_name="telco_linear", data_dir="/home/cdsw")
# Creates an explained version of a partiuclar data point. This is almost exactly the same as the data used in the model serving code.
def explainid(N):
customer_data = dataid(N)[0]
customer_data.pop("id")
customer_data.pop("Churn probability")
data = em.cast_dct(customer_data)
probability, explanation = em.explain_dct(data)
return {
"data": dict(data),
"probability": probability,
"explanation": explanation,
"id": int(N),
}