提交 28e9f9fe authored 作者: James Bergstra's avatar James Bergstra

ENH: Feature base class in gof.fg

上级 5a523b34
......@@ -19,6 +19,7 @@ class InconsistencyError(Exception):
"""
pass
class MissingInputError(Exception):
"""
A symbolic input needed to compute the outputs is missing.
......@@ -26,66 +27,80 @@ class MissingInputError(Exception):
pass
class Feature(object):
"""
Base class for FunctionGraph extensions.
class FunctionGraph(utils.object2):
""" WRITEME
A FunctionGraph represents a subgraph bound by a set of input variables and a
set of output variables, ie a subgraph that specifies a theano function.
The inputs list should contain all the inputs
on which the outputs depend. Variables of type Constant are
not counted as inputs.
The FunctionGraph supports the replace operation which allows to replace a
variable in the subgraph by another, e.g. replace (x + x).out by (2
* x).out. This is the basis for optimization in theano.
This class is also reponsible for verifying that a graph is valid
(ie, all the dtypes and broadcast patterns are compatible with the
way the the Variables are used) and for annotating the Variables with
a .clients field that specifies which Apply nodes use the variable.
The .clients field combined with the .owner field and the Apply nodes'
.inputs field allows the graph to be traversed in both directions.
It can also be "extended" using function_graph.extend(some_object). See the
toolbox and ext modules for common extensions.
Features added with the`extend` function can handle the following events:
See toolbox and ext modules for common extensions.
"""
- feature.on_attach(function_graph)
def on_attach(self, function_graph):
"""
Called by extend. The feature has great freedom in what
it can do with the function_graph: it may, for example, add methods
to it dynamically.
"""
- feature.on_detach(function_graph)
def on_detach(self, function_graph):
"""
Called by remove_feature(feature). Should remove any dynamically-added
functionality that it installed into the function_graph.
"""
- feature.on_import(function_graph, node)*
def on_import(self, function_graph, node):
"""
Called whenever a node is imported into function_graph, which is
just before the node is actually connected to the graph.
"""
- feature.on_prune(function_graph, node)*
def on_prune(self, function_graph, node):
"""
Called whenever a node is pruned (removed) from the function_graph,
after it is disconnected from the graph.
"""
- feature.on_change_input(function_graph, node, i, r, new_r, [reason=None])*
def on_change_input(self, function_graph, node, i, r, new_r, reason=None):
"""
Called whenever node.inputs[i] is changed from r to new_r.
At the moment the callback is done, the change has already
taken place.
- feature.orderings(function_graph)
If you raise an exception in this function, the state of the graph
might be broken for all intents and purposes.
"""
def orderings(self, function_graph):
"""
Called by toposort. It should return a dictionary of
{node: predecessors} where predecessors is a list of
nodes that should be computed before the key node.
* If you raise an exception in the functions marked with an
asterisk, the state of the graph might be inconsistent.
If you raise an exception in this function, the state of the graph
might be broken for all intents and purposes.
"""
- feature.on_setup_node(function_graph, node):
WRITEME
class FunctionGraph(utils.object2):
""" WRITEME
A FunctionGraph represents a subgraph bound by a set of input variables and a
set of output variables, ie a subgraph that specifies a theano function.
The inputs list should contain all the inputs
on which the outputs depend. Variables of type Constant are
not counted as inputs.
The FunctionGraph supports the replace operation which allows to replace a
variable in the subgraph by another, e.g. replace (x + x).out by (2
* x).out. This is the basis for optimization in theano.
This class is also reponsible for verifying that a graph is valid
(ie, all the dtypes and broadcast patterns are compatible with the
way the the Variables are used) and for annotating the Variables with
a .clients field that specifies which Apply nodes use the variable.
The .clients field combined with the .owner field and the Apply nodes'
.inputs field allows the graph to be traversed in both directions.
- feature.on_setup_variable(function_graph, variable):
WRITEME
It can also be "extended" using function_graph.extend(some_object).
See Feature above for event types and documentation.
Historically, the FunctionGraph was called an Env. Keep this in mind
while reading out-of-date documentation, e-mail support threads, etc.
......
Markdown 格式
0%
您添加了 0 到此讨论。请谨慎行事。
请先完成此评论的编辑!
注册 或者 后发表评论