caffe2.python.layer_test_util.LayersTestCase Class Reference

Inheritance diagram for caffe2.python.layer_test_util.LayersTestCase:

Public Member Functions
def	setUp (self)
def	setup_example (self)
def	reset_model (self, input_feature_schema=None, trainer_extra_schema=None)
def	new_record (self, schema_obj)
def	get_training_nets (self, add_constants=False)
def	get_eval_net (self)
def	get_predict_net (self)
def	run_train_net (self)
def	run_train_net_forward_only (self, num_iter=1)
def	assertBlobsEqual (self, spec_blobs, op_blobs)
def	assertArgsEqual (self, spec_args, op_args)
def	assertNetContainOps (self, net, op_specs)
Public Member Functions inherited from caffe2.python.test_util.TestCase
def	setUpClass (cls)
def	setUp (self)
def	tearDown (self)

Public Attributes
	model
Public Attributes inherited from caffe2.python.test_util.TestCase
	ws

Detailed Description

Definition at line 30 of file layer_test_util.py.

Member Function Documentation

def caffe2.python.layer_test_util.LayersTestCase.assertBlobsEqual	(		self,
			spec_blobs,
			op_blobs
	)

spec_blobs can either be None or a list of blob names. If it's None,
then no assertion is performed. The elements of the list can be None,
in that case, it means that position will not be checked.

Definition at line 96 of file layer_test_util.py.

def caffe2.python.layer_test_util.LayersTestCase.assertNetContainOps	(		self,
			net,
			op_specs
	)

Given a net and a list of OpSpec's, check that the net match the spec

Definition at line 125 of file layer_test_util.py.

def caffe2.python.layer_test_util.LayersTestCase.get_training_nets	(		self,
			add_constants = False
	)

We don't use
layer_model_instantiator.generate_training_nets_forward_only()
here because it includes initialization of global constants, which make
testing tricky

Definition at line 57 of file layer_test_util.py.

def caffe2.python.layer_test_util.LayersTestCase.setup_example

(

self

)

This is undocumented feature in hypothesis,
https://github.com/HypothesisWorks/hypothesis-python/issues/59

Definition at line 36 of file layer_test_util.py.

---

The documentation for this class was generated from the following file:

• caffe2/python/layer_test_util.py