PROSAGA码农传奇-深度学习-TensorFlow自己保存一个SparseTensor

<div class =“post-text”itemprop =“text”>
  <P>
    您可以尝试制作TensorFlow常量并将它们存储在一个
     <code>
      GraphDef
    </code>
     文件，然后加载它并在需要时将它们导入图表。我不知道这是否会比你现在的方法更快。
  </p>
  <P>
    要将常量导出到文件，您可以执行以下操作：
  </p>
   <pre>
    <code>
      import tensorflow as tf

# In an independent graph to make sure only the data we want is stored
with tf.Graph().as_default():
    sparse = coo_matrix(dense)
    sparse_indicies = list(zip(
        sparse.row.astype(np.int64).tolist(),
        sparse.col.astype(np.int64).tolist()
    ))
    type_casted = (sparse.data).astype(np.float32)
    # Make TensorFlow constants
    indices = tf.constant(sparse_indicies, name='Indices', dtype=tf.int64)
    values = tf.constant(type_casted, name='Values')
    shape = tf.constant(sparse.shape, dtype=tf.int64, name='Shape')
    # Serialize graph
    graph_def = tf.get_default_graph().as_graph_def()
    with open('sparse_tensor_data.pb', 'wb') as f:
        f.write(graph_def.SerializeToString())

</code>
  </pre>
  <P>
    您可以从其他地方导入它，如下所示：
  </p>
   <pre>
    <code>
      import tensorflow as tf

# Read graph
graph_def = tf.GraphDef()
with open('sparse_tensor_data.pb', 'rb') as f:
    graph_def.MergeFromString(f.read())
# Import tensors
indices, values, shape = tf.import_graph_def(
    graph_def, return_elements=['Indices:0', 'Values:0', 'Shape:0'], name='SparseTensorImport')
del graph_def
# Create sparse tensor
input_tensor = tf.SparseTensor(indices=indices, values=values, dense_shape=shape)

</code>
  </pre>
</DIV>