waytrue17 commented on a change in pull request #19543:
URL: https://github.com/apache/incubator-mxnet/pull/19543#discussion_r525539102
##########
File path: python/mxnet/gluon/block.py
##########
@@ -1259,45 +1282,55 @@ def register_child(self, block, name=None):
self._active = False
self._clear_cached_op()
- def hybridize(self, active=True, backend=None, backend_opts=None,
clear=True, partition_if_dynamic=False, **kwargs):
+ def hybridize(self, active=True,
+ partition_if_dynamic=False,
+ static_alloc=False,
+ static_shape=False,
+ inline_limit=2,
+ forward_bulk_size=None,
+ backward_bulk_size=None):
"""Activates or deactivates :py:class:`HybridBlock` s recursively. Has
no effect on
non-hybrid children.
Parameters
----------
active : bool, default True
Whether to turn hybrid on or off.
- backend : str
- The name of backend, as registered in `SubgraphBackendRegistry`,
default None
- backend_opts : dict of user-specified options to pass to the backend
for partitioning, optional
- Passed on to `PrePartition` and `PostPartition` functions of
`SubgraphProperty`
- clear : clears any previous optimizations
- partition_if_dynamic : bool
+ partition_if_dynamic : bool, default False
whether to partition the graph when dynamic shape op exists
static_alloc : bool, default False
Statically allocate memory to improve speed. Memory usage may
increase.
static_shape : bool, default False
Optimize for invariant input shapes between iterations. Must also
set static_alloc to True. Change of input shapes is still allowed
but slower.
+ inline_limit : optional int, default 2
+ Maximum number of operators that can be inlined.
+ forward_bulk_size : optional int, default None
+ Segment size of bulk execution during forward pass.
+ backward_bulk_size : optional int, default None
+ Segment size of bulk execution during forward pass.
Review comment:
Same here
##########
File path: example/extensions/lib_pass/README.md
##########
@@ -83,17 +84,7 @@ APIs in MXNet are available in both Symbol and Gluon APIs.
For the Symbol API, `
sym.optimize_for(backend, args=None, aux=None, ctx=None, **kwargs)
```
-The `optimize_for` API takes at least 1 argument, `backend` which is a string
that identifies which backend to use to optimize the model. The `args` and
`aux` arguments are optional and take a list of NDArray or dict of str to
NDArray. They are used to infer shapes and types and before executing the graph
pass. The `ctx` argument is optional and takes a device context to infer
storage types. It also takes any other user-specified options that will be
passed to the backend APIs.
-
-For the Gluon API, `hybridize` can be called on HybridBlocks to execute a
graph pass on the internal CachedOp Symbol.
-
-```python
-block.hybridize(backend=None, backend_opts=None, **kwargs)
-```
-
-The `hybridize` function prepares the HybridBlock to be converted into a
backend symbol. The `backend` argument is a string that identifies which pass
that will be executed on the model. The `backend_opts` takes other
user-specified options that will be passed to the backend APIs. The actual pass
runs once just before the first the forward pass.
-
-If you just want to run a graph pass on the HybridBlock but not run a complete
forward pass, you can use the `optimize_for` API that combines the work done in
the `hybridize` API with part of the work done in the forward pass.
+The `optimize_for` API takes at least 1 argument, `backend` which is a string
that identifies which backend to use to optimize the model. The `args` and
`aux` arguments are optional and take a list of NDArray or dict of str to
NDArray. They are used to infer shapes and types and before executing the graph
pass. The `ctx` argument is optional and takes a device context to infer
storage types. It also takes any other user-specified options that will be
passed to the backend APIs (in the `kwargs`).
Review comment:
Does `optimize_for` take at least 2 arguments, `x` and `backend`?
##########
File path: python/mxnet/gluon/block.py
##########
@@ -1205,19 +1212,32 @@ def optimize_for(self, x, *args, backend=None,
backend_opts=None, clear=True, pa
The name of backend, as registered in `SubgraphBackendRegistry`,
default None
backend_opts : dict of user-specified options to pass to the backend
for partitioning, optional
Passed on to `PrePartition` and `PostPartition` functions of
`SubgraphProperty`
- clear : clears any previous optimizations
- partition_if_dynamic : bool
+ clear : bool, default False
+ clears any previous optimizations
+ partition_if_dynamic : bool, default False
whether to partition the graph when dynamic shape op exists
static_alloc : bool, default False
Statically allocate memory to improve speed. Memory usage may
increase.
static_shape : bool, default False
Optimize for invariant input shapes between iterations. Must also
set static_alloc to True. Change of input shapes is still allowed
but slower.
+ inline_limit : optional int, default 2
+ Maximum number of operators that can be inlined.
+ forward_bulk_size : optional int, default None
+ Segment size of bulk execution during forward pass.
+ backward_bulk_size : optional int, default None
+ Segment size of bulk execution during forward pass.
Review comment:
Should this be "during backward pass"?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
