def __init__(self,
vectorizer=None,
model=None,
x_train=None,
y_train=None,
x_test=None,
y_test=None,
feature_names=None,
metadata=None,
*args,
**kwargs):
if model:
self._vectorizer = vectorizer
self._model = model
self._data = {
'x_train': x_train,
'x_train_vec': vectorizer.fit_transform(x_test),
'y_train': y_train,
'x_test': x_test,
'x_test_vec': vectorizer.transform(x_test),
'y_test': y_test,
}
now = datetime.datetime.now().strftime('%Y-%m-%d-%H-%M-%S')
if metadata:
self.metadata = metadata
else:
pred = model.predict(self._data['x_test_vec'])
precision, recall, fscore, support = precision_recall_fscore_support(
y_test, pred)
fpr, tpr, thresholds = roc_curve(y_test, pred)
auc_score = auc(fpr, tpr)
self.metadata = {
'vectorizer': vectorizer_to_str(vectorizer),
'model': str(model),
'created_at': now,
'feature_names': feature_names,
'git_hash': get_git_hash(),
'precision': [float(p) for p in precision],
'recall': [float(r) for r in recall],
'fscore': [float(f) for f in fscore],
'support': [int(s) for s in support],
'auc': auc_score
}
log.info('Initialized PackagedModel %s' % now)
def __init__(self,
pipeline=None,
feature_union=None,
x_train=None,
y_train=None,
x_test=None,
y_test=None,
metadata=None,
*args,
**kwargs):
if pipeline:
self._feature_union = feature_union # will this maintain state?
self._pipeline = pipeline
self._data = {
'x_train': x_train,
'y_train': y_train,
'x_test': x_test,
'y_test': y_test,
}
now = datetime.datetime.now().strftime('%Y%m%d%H%M%S')
if metadata:
self.metadata = metadata
else:
x_test_vec = feature_union.transform(self._data['x_test'])
pred = pipeline.predict(x_test_vec)
precision, recall, fscore, support = precision_recall_fscore_support(
y_test, pred)
fpr, tpr, thresholds = roc_curve(y_test, pred)
auc_score = auc(fpr, tpr)
self.metadata = {
'pipeline': str(pipeline),
'feature_union': str(feature_union),
'created_at': now,
'git_hash': get_git_hash(),
'precision': [float(p) for p in precision],
'recall': [float(r) for r in recall],
'fscore': [float(f) for f in fscore],
'support': [int(s) for s in support],
'auc': auc_score
}
log.info('Initialized PackagedPipeline %s' % now)
def __init__(
self,
vectorizer=None, model=None,
x_train=None, y_train=None, x_test=None, y_test=None,
feature_names=None, metadata=None,
*args, **kwargs
):
if model:
self._vectorizer = vectorizer
self._model = model
self._data = {
'x_train': x_train,
'x_train_vec': vectorizer.fit_transform(x_test),
'y_train': y_train,
'x_test': x_test,
'x_test_vec': vectorizer.transform(x_test),
'y_test': y_test,
}
now = datetime.datetime.now().strftime('%Y-%m-%d-%H-%M-%S')
if metadata:
self.metadata = metadata
else:
pred = model.predict(self._data['x_test_vec'])
precision, recall, fscore, support = precision_recall_fscore_support(y_test, pred)
fpr, tpr, thresholds = roc_curve(y_test, pred)
auc_score = auc(fpr, tpr)
self.metadata = {
'vectorizer': vectorizer_to_str(vectorizer),
'model': str(model),
'created_at': now,
'feature_names': feature_names,
'git_hash': get_git_hash(),
'precision': [float(p) for p in precision],
'recall': [float(r) for r in recall],
'fscore': [float(f) for f in fscore],
'support': [int(s) for s in support],
'auc': auc_score
}
log.info('Initialized PackagedModel %s' % now)
def __init__(
self,
pipeline=None, feature_union=None,
x_train=None, y_train=None, x_test=None, y_test=None,
metadata=None,
*args, **kwargs
):
if pipeline:
self._feature_union = feature_union # will this maintain state?
self._pipeline = pipeline
self._data = {
'x_train': x_train,
'y_train': y_train,
'x_test': x_test,
'y_test': y_test,
}
now = datetime.datetime.now().strftime('%Y%m%d%H%M%S')
if metadata:
self.metadata = metadata
else:
x_test_vec = feature_union.transform(self._data['x_test'])
pred = pipeline.predict(x_test_vec)
precision, recall, fscore, support = precision_recall_fscore_support(y_test, pred)
fpr, tpr, thresholds = roc_curve(y_test, pred)
auc_score = auc(fpr, tpr)
self.metadata = {
'pipeline': str(pipeline),
'feature_union': str(feature_union),
'created_at': now,
'git_hash': get_git_hash(),
'precision': [float(p) for p in precision],
'recall': [float(r) for r in recall],
'fscore': [float(f) for f in fscore],
'support': [int(s) for s in support],
'auc': auc_score
}
log.info('Initialized PackagedPipeline %s' % now)
