From 3dfbef290b98d30ac7f1f94da07e07f52dc41374 Mon Sep 17 00:00:00 2001
From: JiabinYang <marsyang199376@gmail.com>
Date: Mon, 28 Jan 2019 06:17:42 +0000
Subject: [PATCH] polish code and add comments for Embedding

---
 python/paddle/fluid/imperative/nn.py          | 45 ++++++++++++++++---
 .../unittests/test_imperative_ptb_rnn.py      | 11 ++---
 2 files changed, 46 insertions(+), 10 deletions(-)

diff --git a/python/paddle/fluid/imperative/nn.py b/python/paddle/fluid/imperative/nn.py
index b5c049e927e..ea04475493f 100644
--- a/python/paddle/fluid/imperative/nn.py
+++ b/python/paddle/fluid/imperative/nn.py
@@ -22,7 +22,7 @@ from . import layers
 from ..framework import Variable, OpProtoHolder
 from ..param_attr import ParamAttr
 from ..initializer import Normal, Constant
-__all__ = ['Conv2D', 'Pool2D', 'FC', 'BatchNorm', 'EMBEDDING']
+__all__ = ['Conv2D', 'Pool2D', 'FC', 'BatchNorm', 'Embedding']
 
 
 class Conv2D(layers.Layer):
@@ -415,7 +415,44 @@ class BatchNorm(layers.Layer):
         return self._helper.append_activation(batch_norm_out)
 
 
-class EMBEDDING(layers.Layer):
+class Embedding(layers.Layer):
+    """
+    **Embedding Layer**
+
+    This layer is used to lookup embeddings of IDs, provided by :attr:`input`, in
+    a lookup table. The result of this lookup is the embedding of each ID in the
+    :attr:`input`.
+
+    All the input variables are passed in as local variables to the LayerHelper
+    constructor.
+
+    Args:
+        size(tuple|list): The shape of the look up table parameter. It should
+            have two elements which indicate the size of the dictionary of
+            embeddings and the size of each embedding vector respectively.
+        is_sparse(bool): The flag indicating whether to use sparse update.
+        is_distributed(bool): Whether to run lookup table from remote parameter server.
+        padding_idx(int|long|None): If :attr:`None`, it makes no effect to lookup.
+            Otherwise the given :attr:`padding_idx` indicates padding the output
+            with zeros whenever lookup encounters it in :attr:`input`. If
+            :math:`padding_idx < 0`, the :attr:`padding_idx` to use in lookup is
+            :math:`size[0] + dim`.
+        param_attr(ParamAttr): Parameters for this layer
+        dtype(np.dtype|core.VarDesc.VarType|str): The type of data : float32, float_16, int etc
+
+    Returns:
+        Variable: The tensor variable storing the embeddings of the \
+                  supplied inputs.
+
+    Examples:
+        .. code-block:: python
+
+          dict_size = len(dataset.ids)
+          input = fluid.layers.data(name='ids', shape=[32, 32], dtype='float32')
+          embedding = fluid.imperative.Embedding(size=[dict_size, 16])
+          fc = embedding(input)
+    """
+
     def __init__(self,
                  size,
                  is_sparse=False,
@@ -424,7 +461,7 @@ class EMBEDDING(layers.Layer):
                  param_attr=None,
                  dtype='float32'):
 
-        super(EMBEDDING, self).__init__()
+        super(Embedding, self).__init__()
         self._size = size
         self._is_sparse = is_sparse
         self._is_distributed = is_distributed
@@ -440,8 +477,6 @@ class EMBEDDING(layers.Layer):
 
         from ..layer_helper import LayerHelper
         self._helper = LayerHelper('embedding', param_attr=param_attr)
-
-    def _build_once(self, input):
         self._w = self._helper.create_parameter(
             attr=self._param_attr,
             shape=self._size,
diff --git a/python/paddle/fluid/tests/unittests/test_imperative_ptb_rnn.py b/python/paddle/fluid/tests/unittests/test_imperative_ptb_rnn.py
index 9c6ec331e6e..a3e3f967131 100644
--- a/python/paddle/fluid/tests/unittests/test_imperative_ptb_rnn.py
+++ b/python/paddle/fluid/tests/unittests/test_imperative_ptb_rnn.py
@@ -16,7 +16,7 @@ from __future__ import print_function
 
 import unittest
 import paddle.fluid as fluid
-from paddle.fluid.imperative.nn import EMBEDDING
+from paddle.fluid.imperative.nn import Embedding
 import paddle.fluid.framework as framework
 from paddle.fluid.optimizer import SGDOptimizer
 from paddle.fluid.imperative.base import to_variable
@@ -143,7 +143,7 @@ class PtbModel(fluid.imperative.Layer):
             num_layers=num_layers,
             init_scale=init_scale,
             dropout=dropout)
-        self.embedding = EMBEDDING(
+        self.embedding = Embedding(
             size=[vocab_size, hidden_size],
             dtype='float32',
             is_sparse=False,
@@ -151,8 +151,6 @@ class PtbModel(fluid.imperative.Layer):
                 name='embedding_para',
                 initializer=fluid.initializer.UniformInitializer(
                     low=-init_scale, high=init_scale)))
-
-    def _build_once(self, input, label, init_hidden, init_cell):
         self.softmax_weight = fluid.layers.create_parameter(
             [self.hidden_size, self.vocab_size],
             dtype="float32",
@@ -166,6 +164,9 @@ class PtbModel(fluid.imperative.Layer):
             default_initializer=fluid.initializer.UniformInitializer(
                 low=-self.init_scale, high=self.init_scale))
 
+    def _build_once(self, input, label, init_hidden, init_cell):
+        pass
+
     def forward(self, input, label, init_hidden, init_cell):
 
         init_h = fluid.layers.reshape(
@@ -203,7 +204,7 @@ class PtbModel(fluid.imperative.Layer):
 
 
 class TestImperativePtbRnn(unittest.TestCase):
-    def test_mnist_cpu_float32(self):
+    def test_ptb_rnn_cpu_float32(self):
         seed = 90
         hidden_size = 10
         vocab_size = 1000
-- 
GitLab