diff --git a/python/paddle/v2/layer.py b/python/paddle/v2/layer.py index a91c6586a48aae4c692088e52d179efe4f69dbdd..141f7b359196499ee61e6151320fd75dc8b78c31 100644 --- a/python/paddle/v2/layer.py +++ b/python/paddle/v2/layer.py @@ -53,20 +53,29 @@ import data_type __all__ = ['parse_network', 'data'] -def parse_network(*outputs, **kwargs): +def parse_network(output_layers, extra_layers=None): """ - Parse all output layers and then generate a ModelConfig object. + Parse all layers in the neural network graph and + then generate a ModelConfig object. .. note:: This function is used internally in paddle.v2 module. User should never invoke this method. - :param outputs: Output layers. - :type outputs: Layer + :param output_layers: Output layers. + :type output_layers: Layer + :param extra_layers: Some layers in the neural network graph are not in the + path of output_layers. + :type extra_layers: Layer :return: A ModelConfig object instance. :rtype: ModelConfig """ + if not isinstance(output_layers, collections.Sequence): + output_layers = [output_layers] + if extra_layers is not None and not isinstance(extra_layers, + collections.Sequence): + extra_layers = [extra_layers] def __real_func__(): """ @@ -74,8 +83,7 @@ def parse_network(*outputs, **kwargs): the plain old paddle configuration function. """ context = dict() - real_output = [each.to_proto(context=context) for each in outputs] - extra_layers = kwargs.get('extra_layers', None) + real_output = [each.to_proto(context=context) for each in output_layers] if extra_layers is not None: extra_output = [ each.to_proto(context=context) for each in extra_layers diff --git a/python/paddle/v2/tests/test_layer.py b/python/paddle/v2/tests/test_layer.py index 2378f68ea89aa0268eb2482359e1fde71442258d..c67f3b84d96eb92d94ad80cc54c5e056103c1a1a 100644 --- a/python/paddle/v2/tests/test_layer.py +++ b/python/paddle/v2/tests/test_layer.py @@ -59,13 +59,13 @@ class ImageLayerTest(unittest.TestCase): num_channels=16, pool_type=pooling.Max()) maxout = layer.maxout(input=conv, num_channels=16, groups=4) - print layer.parse_network(maxpool, spp, maxout) + print layer.parse_network([maxpool, spp, maxout]) def test_norm_layer(self): norm1 = layer.img_cmrnorm(input=conv, size=5) norm2 = layer.batch_norm(input=conv) norm3 = layer.sum_to_one_norm(input=conv) - print layer.parse_network(norm1, norm2, norm3) + print layer.parse_network([norm1, norm2, norm3]) class AggregateLayerTest(unittest.TestCase): @@ -78,7 +78,8 @@ class AggregateLayerTest(unittest.TestCase): first_seq = layer.first_seq(input=pixel) concat = layer.concat(input=[last_seq, first_seq]) seq_concat = layer.seq_concat(a=last_seq, b=first_seq) - print layer.parse_network(pool, last_seq, first_seq, concat, seq_concat) + print layer.parse_network( + [pool, last_seq, first_seq, concat, seq_concat]) class MathLayerTest(unittest.TestCase): @@ -95,8 +96,10 @@ class MathLayerTest(unittest.TestCase): tensor = layer.tensor(a=pixel, b=pixel, size=1000) cos_sim = layer.cos_sim(a=pixel, b=pixel) trans = layer.trans(input=tensor) - print layer.parse_network(addto, linear_comb, interpolation, power, - scaling, slope, tensor, cos_sim, trans) + print layer.parse_network([ + addto, linear_comb, interpolation, power, scaling, slope, tensor, + cos_sim, trans + ]) class ReshapeLayerTest(unittest.TestCase): @@ -110,7 +113,8 @@ class ReshapeLayerTest(unittest.TestCase): repeat = layer.repeat(input=pixel, num_repeats=4) reshape = layer.seq_reshape(input=pixel, reshape_size=4) rotate = layer.rotate(input=pixel, height=16, width=49) - print layer.parse_network(block_expand, expand, repeat, reshape, rotate) + print layer.parse_network( + [block_expand, expand, repeat, reshape, rotate]) class RecurrentLayerTest(unittest.TestCase): @@ -119,7 +123,7 @@ class RecurrentLayerTest(unittest.TestCase): recurrent = layer.recurrent(input=word) lstm = layer.lstmemory(input=word) gru = layer.grumemory(input=word) - print layer.parse_network(recurrent, lstm, gru) + print layer.parse_network([recurrent, lstm, gru]) class CostLayerTest(unittest.TestCase): @@ -139,10 +143,10 @@ class CostLayerTest(unittest.TestCase): cost10 = layer.sum_cost(input=inference) cost11 = layer.huber_cost(input=score, label=label) - print layer.parse_network(cost1, cost2) - print layer.parse_network(cost3, cost4) - print layer.parse_network(cost5, cost6) - print layer.parse_network(cost7, cost8, cost9, cost10, cost11) + print layer.parse_network([cost1, cost2]) + print layer.parse_network([cost3, cost4]) + print layer.parse_network([cost5, cost6]) + print layer.parse_network([cost7, cost8, cost9, cost10, cost11]) crf = layer.crf(input=inference, label=label) crf_decoding = layer.crf_decoding(input=inference, size=3) @@ -151,8 +155,8 @@ class CostLayerTest(unittest.TestCase): nce = layer.nce(input=inference, label=label, num_classes=3) hsigmoid = layer.hsigmoid(input=inference, label=label, num_classes=3) - print layer.parse_network(crf, crf_decoding, ctc, warp_ctc, nce, - hsigmoid) + print layer.parse_network( + [crf, crf_decoding, ctc, warp_ctc, nce, hsigmoid]) class OtherLayerTest(unittest.TestCase): @@ -160,7 +164,7 @@ class OtherLayerTest(unittest.TestCase): maxid = layer.max_id(input=inference) sampling_id = layer.sampling_id(input=inference) eos = layer.eos(input=maxid, eos_id=5) - print layer.parse_network(maxid, sampling_id, eos) + print layer.parse_network([maxid, sampling_id, eos]) def test_slicing_joining_layer(self): pad = layer.pad(input=conv, pad_c=[2, 3], pad_h=[1, 2], pad_w=[3, 1]) diff --git a/python/paddle/v2/topology.py b/python/paddle/v2/topology.py index dbcd86b54fd2bee001efaf482721710e2a880496..07e0fb74289b649653b42488f0e1c65af8430f72 100644 --- a/python/paddle/v2/topology.py +++ b/python/paddle/v2/topology.py @@ -13,7 +13,6 @@ # limitations under the License. import collections -import copy from paddle.proto.ModelConfig_pb2 import ModelConfig @@ -69,7 +68,7 @@ class Topology(object): extra_layers = __check__(extra_layers) self.__model_config__ = v2_layer.parse_network( - *layers, extra_layers=extra_layers) + layers, extra_layers=extra_layers) if extra_layers is not None: self.layers.extend(extra_layers) diff --git a/python/paddle/v2/trainer.py b/python/paddle/v2/trainer.py index cccb00794acdf936d2e11e68ae1573f99a8a5155..a207beb548f0fda6d7aacf5c55e53d937e18a924 100644 --- a/python/paddle/v2/trainer.py +++ b/python/paddle/v2/trainer.py @@ -37,6 +37,9 @@ class SGD(object): :type cost: paddle.v2.config_base.Layer :param parameters: The parameters dictionary. :type parameters: paddle.v2.parameters.Parameters + :param extra_layers: Some layers in the neural network graph are not + in the path of cost layer. + :type extra_layers: paddle.v2.config_base.Layer """ def __init__(self, cost, parameters, update_equation, extra_layers=None):