def testTrackPersistentBytes(self):
ops.reset_default_graph()
a = array_ops.constant(np.ones((100, 100)))
b = array_ops.constant(np.ones((100, 100)))
c = a * b
with session.Session() as sess:
run_options = config_pb2.RunOptions(
trace_level=config_pb2.RunOptions.FULL_TRACE)
run_metadata = config_pb2.RunMetadata()
sess.run(c, options=run_options, run_metadata=run_metadata)
options = option_builder.ProfileOptionBuilder.time_and_memory()
options['min_bytes'] = 0
options['select'] = ('bytes', 'peak_bytes', 'output_bytes',
'residual_bytes')
ret = model_analyzer.profile(
sess.graph, run_meta=run_metadata, cmd='scope', options=options)
run_metadata = config_pb2.RunMetadata()
sess.run(c, options=run_options, run_metadata=run_metadata)
ret2 = model_analyzer.profile(
sess.graph, run_meta=run_metadata, cmd='scope', options=options)
n = lib.SearchTFProfNode(ret, 'mul')
n2 = lib.SearchTFProfNode(ret2, 'mul')
self.assertGreater(n.peak_bytes, 0)
self.assertGreater(n.output_bytes, 0)
self.assertGreater(n.residual_bytes, 0)
self.assertEqual(n.peak_bytes, n2.peak_bytes)
self.assertEqual(n.output_bytes, n2.output_bytes)
self.assertEqual(n.residual_bytes, n2.residual_bytes)
def testMultipleProfilePerStep(self):
ops.reset_default_graph()
opts = (builder(builder.trainable_variables_parameter())
.with_empty_output()
.with_accounted_types(['.*'])
.select(['micros', 'bytes', 'peak_bytes',
'residual_bytes', 'output_bytes']).build())
r = lib.BuildSmallModel()
sess = session.Session()
profiler = model_analyzer.Profiler(sess.graph)
init_var_run_meta = config_pb2.RunMetadata()
sess.run(variables.global_variables_initializer(),
options=config_pb2.RunOptions(
trace_level=config_pb2.RunOptions.FULL_TRACE),
run_metadata=init_var_run_meta)
train_run_meta = config_pb2.RunMetadata()
sess.run(r,
options=config_pb2.RunOptions(
trace_level=config_pb2.RunOptions.FULL_TRACE),
run_metadata=train_run_meta)
profiler.add_step(0, train_run_meta)
ret1 = profiler.profile_name_scope(opts)
n1 = lib.SearchTFProfNode(
ret1, 'DW/Initializer/random_normal/RandomStandardNormal')
# Without the var initialization run_meta, it doesn't have the
# information of var_initialization.
self.assertEqual(n1.exec_micros, 0)
self.assertEqual(n1.requested_bytes, 0)
self.assertEqual(n1.peak_bytes, 0)
self.assertEqual(n1.residual_bytes, 0)
profiler.add_step(0, init_var_run_meta)
ret2 = profiler.profile_name_scope(opts)
n2 = lib.SearchTFProfNode(
ret2, 'DW/Initializer/random_normal/RandomStandardNormal')
# After adding the var initialization run_meta.
self.assertGreater(n2.exec_micros, 0)
self.assertGreater(n2.requested_bytes, 0)
self.assertGreater(n2.peak_bytes, 0)
self.assertGreater(n2.residual_bytes, 0)
def testLoopCPU(self):
ops.reset_default_graph()
with ops.device('/cpu:0'):
tfprof_node, run_meta = _run_loop_model()
# The while-loop caused a node to appear 4 times in scheduling.
ret = _extract_node(run_meta, 'rnn/while/basic_rnn_cell/MatMul')
self.assertEqual(len(ret['cpu:0']), 4)
total_cpu_execs = 0
for node in ret['cpu:0']:
total_cpu_execs += node.op_end_rel_micros
mm_node = lib.SearchTFProfNode(tfprof_node,
'rnn/while/basic_rnn_cell/MatMul')
self.assertEqual(mm_node.run_count, 4)
self.assertEqual(mm_node.cpu_exec_micros, total_cpu_execs)
self.assertEqual(mm_node.exec_micros, total_cpu_execs)
def testMultiStepProfile(self):
ops.reset_default_graph()
opts = builder.time_and_memory(min_bytes=0)
with session.Session() as sess:
r1, r2, r3 = lib.BuildSplittableModel()
sess.run(variables.global_variables_initializer())
profiler = model_analyzer.Profiler(sess.graph)
pb0 = profiler.profile_name_scope(opts)
run_meta = config_pb2.RunMetadata()
_ = sess.run(r1,
options=config_pb2.RunOptions(
trace_level=config_pb2.RunOptions.FULL_TRACE),
run_metadata=run_meta)
profiler.add_step(1, run_meta)
pb1 = profiler.profile_name_scope(opts)
self.assertNotEqual(lib.SearchTFProfNode(pb1, 'DW'), None)
self.assertEqual(lib.SearchTFProfNode(pb1, 'DW2'), None)
self.assertEqual(lib.SearchTFProfNode(pb1, 'add'), None)
run_meta2 = config_pb2.RunMetadata()
_ = sess.run(r2,
options=config_pb2.RunOptions(
trace_level=config_pb2.RunOptions.FULL_TRACE),
run_metadata=run_meta2)
profiler.add_step(2, run_meta2)
pb2 = profiler.profile_name_scope(opts)
self.assertNotEqual(lib.SearchTFProfNode(pb2, 'DW'), None)
self.assertNotEqual(lib.SearchTFProfNode(pb2, 'DW2'), None)
self.assertEqual(lib.SearchTFProfNode(pb2, 'add'), None)
run_meta3 = config_pb2.RunMetadata()
_ = sess.run(r3,
options=config_pb2.RunOptions(
trace_level=config_pb2.RunOptions.FULL_TRACE),
run_metadata=run_meta3)
profiler.add_step(3, run_meta3)
pb3 = profiler.profile_name_scope(opts)
self.assertNotEqual(lib.SearchTFProfNode(pb3, 'DW'), None)
self.assertNotEqual(lib.SearchTFProfNode(pb3, 'DW2'), None)
self.assertNotEqual(lib.SearchTFProfNode(pb3, 'add'), None)
self.assertEqual(lib.SearchTFProfNode(pb0, 'Conv2D'), None)
self.assertGreater(lib.SearchTFProfNode(pb1, 'Conv2D').exec_micros, 0)
self.assertEqual(lib.SearchTFProfNode(pb1, 'Conv2D_1'), None)
self.assertGreater(lib.SearchTFProfNode(pb2, 'Conv2D_1').exec_micros, 0)
self.assertEqual(lib.SearchTFProfNode(pb2, 'add'), None)
self.assertGreater(lib.SearchTFProfNode(pb3, 'add').exec_micros, 0)
advice_pb = profiler.advise(model_analyzer.ALL_ADVICE)
self.assertTrue('AcceleratorUtilizationChecker' in advice_pb.checkers)
self.assertTrue('ExpensiveOperationChecker' in advice_pb.checkers)
self.assertTrue('OperationChecker' in advice_pb.checkers)
checker = advice_pb.checkers['AcceleratorUtilizationChecker']
if test.is_gpu_available():
self.assertGreater(len(checker.reports), 0)
else:
self.assertEqual(len(checker.reports), 0)
checker = advice_pb.checkers['ExpensiveOperationChecker']
self.assertGreater(len(checker.reports), 0)
