LostTech.TensorFlow : API Documentation

Type DenseHashTable

Namespace tensorflow.lookup.experimental

Parent LookupInterface

Interfaces IDenseHashTable

A generic mutable hash table implementation using tensors as backing store.

Data can be inserted by calling the insert method and removed by calling the remove method. It does not support initialization via the init method.

It uses "open addressing" with quadratic reprobing to resolve collisions. Compared to `MutableHashTable` the insert, remove and lookup operations in a `DenseHashTable` are typically faster, but memory usage can be higher. However, `DenseHashTable` does not require additional memory for temporary tensors created during checkpointing and restore operations.

Example usage:
Show Example 
table = tf.lookup.DenseHashTable(key_dtype=tf.int64,
                                             value_dtype=tf.int64,
                                             default_value=-1,
                                             empty_key=0,
                                             deleted_key=-1) 
 sess.run(table.insert(keys, values))
out = table.lookup(query_keys)
print(out.eval())

Methods

Properties

Public instance methods

object erase(IGraphNodeBase keys, string name)

Removes `keys` and its associated values from the table.

If a key is not present in the table, it is silently ignored.
Parameters
IGraphNodeBase keys
Keys to remove. Can be a tensor of any shape. Must match the table's key type.
string name
A name for the operation (optional).
Returns
object
The created Operation.

object erase_dyn(object keys, object name)

Removes `keys` and its associated values from the table.

If a key is not present in the table, it is silently ignored.
Parameters
object keys
Keys to remove. Can be a tensor of any shape. Must match the table's key type.
object name
A name for the operation (optional).
Returns
object
The created Operation.

object insert_or_assign(IGraphNodeBase keys, IGraphNodeBase values, string name)

Associates `keys` with `values`.
Parameters
IGraphNodeBase keys
Keys to insert. Can be a tensor of any shape. Must match the table's key type.
IGraphNodeBase values
Values to be associated with keys. Must be a tensor of the same shape as `keys` and match the table's value type.
string name
A name for the operation (optional).
Returns
object
The created Operation.

object insert_or_assign_dyn(object keys, object values, object name)

Associates `keys` with `values`.
Parameters
object keys
Keys to insert. Can be a tensor of any shape. Must match the table's key type.
object values
Values to be associated with keys. Must be a tensor of the same shape as `keys` and match the table's value type.
object name
A name for the operation (optional).
Returns
object
The created Operation.

object remove(IGraphNodeBase keys, string name)

Removes `keys` and its associated values from the table.

If a key is not present in the table, it is silently ignored.
Parameters
IGraphNodeBase keys
Keys to remove. Can be a tensor of any shape. Must match the table's key type.
string name
A name for the operation (optional).
Returns
object
The created Operation.

object remove_dyn(object keys, object name)

Removes `keys` and its associated values from the table.

If a key is not present in the table, it is silently ignored.
Parameters
object keys
Keys to remove. Can be a tensor of any shape. Must match the table's key type.
object name
A name for the operation (optional).
Returns
object
The created Operation.

Public static methods

DenseHashTable NewDyn(object key_dtype, object value_dtype, object default_value, object empty_key, object deleted_key, object initial_num_buckets, ImplicitContainer<T> name, ImplicitContainer<T> checkpoint)

Creates an empty `DenseHashTable` object.

Creates a table, the type of its keys and values are specified by key_dtype and value_dtype, respectively.
Parameters
object key_dtype
the type of the key tensors.
object value_dtype
the type of the value tensors.
object default_value
The value to use if a key is missing in the table.
object empty_key
the key to use to represent empty buckets internally. Must not be used in insert, remove or lookup operations.
object deleted_key
the key to use to represent deleted buckets internally. Must not be used in insert, remove or lookup operations and be different from the empty_key.
object initial_num_buckets
the initial number of buckets.
ImplicitContainer<T> name
A name for the operation (optional).
ImplicitContainer<T> checkpoint
if True, the contents of the table are saved to and restored from checkpoints. If `shared_name` is empty for a checkpointed table, it is shared using the table node name.
Returns
DenseHashTable
A `DenseHashTable` object.

Public properties

DType key_dtype get;

object key_dtype_dyn get;

object name get;

The name of the table.

object name_dyn get;

The name of the table.

object PythonObject get;

Tensor resource_handle get;

object resource_handle_dyn get;

DType value_dtype get;

object value_dtype_dyn get;