You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by jx...@apache.org on 2018/04/24 22:04:37 UTC
[incubator-mxnet] branch master updated: move parameters to backend
for hybridblock (#10632)
This is an automated email from the ASF dual-hosted git repository.
jxie pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git
The following commit(s) were added to refs/heads/master by this push:
new faffe56 move parameters to backend for hybridblock (#10632)
faffe56 is described below
commit faffe56b450cfc908bf9d29e810e00c0de8aca7b
Author: Eric Junyuan Xie <pi...@users.noreply.github.com>
AuthorDate: Tue Apr 24 15:04:31 2018 -0700
move parameters to backend for hybridblock (#10632)
---
include/mxnet/base.h | 8 ++++
include/mxnet/c_api.h | 7 ++-
include/mxnet/imperative.h | 18 +++++---
python/mxnet/_ctypes/ndarray.py | 17 ++++++-
python/mxnet/gluon/block.py | 90 +++++++++++++++++-------------------
src/c_api/c_api_ndarray.cc | 31 +++++++++++--
src/executor/attach_op_execs_pass.cc | 11 +----
src/executor/graph_executor.cc | 1 -
src/executor/graph_executor.h | 2 -
src/imperative/cached_op.cc | 67 ++++++++++++++++++++++-----
10 files changed, 168 insertions(+), 84 deletions(-)
diff --git a/include/mxnet/base.h b/include/mxnet/base.h
index 783002e..38fd7ed 100644
--- a/include/mxnet/base.h
+++ b/include/mxnet/base.h
@@ -362,6 +362,14 @@ constexpr size_t kMKLDNNAlign = 64;
} // namespace mxnet
+namespace std {
+template<> struct hash<mxnet::Context> {
+ size_t operator()(const mxnet::Context& ctx) const {
+ return (static_cast<size_t>(ctx.dev_type) << 32) | ctx.dev_id;
+ }
+};
+}
+
#include "./tensor_blob.h"
//! \endcond
#endif // MXNET_BASE_H_
diff --git a/include/mxnet/c_api.h b/include/mxnet/c_api.h
index cbc83b2..3f04051 100644
--- a/include/mxnet/c_api.h
+++ b/include/mxnet/c_api.h
@@ -976,9 +976,14 @@ MXNET_DLL int MXCreateCachedOp(SymbolHandle handle, CachedOpHandle *out);
* \brief create cached operator
*/
MXNET_DLL int MXCreateCachedOpEx(SymbolHandle handle,
- int num_params,
+ int num_flags,
const char** keys,
const char** vals,
+ int num_inputs,
+ const char** input_names,
+ int num_params,
+ const char** param_names,
+ NDArrayHandle* params,
CachedOpHandle *out);
/*!
* \brief free cached operator
diff --git a/include/mxnet/imperative.h b/include/mxnet/imperative.h
index d605e9d..758ce85 100644
--- a/include/mxnet/imperative.h
+++ b/include/mxnet/imperative.h
@@ -36,11 +36,11 @@
namespace mxnet {
/*! \brief CachedOp Parameters */
-struct CachedOpParam : public dmlc::Parameter<CachedOpParam> {
+struct CachedOpConfig : public dmlc::Parameter<CachedOpConfig> {
uint32_t inline_limit;
uint32_t forward_bulk_size;
uint32_t backward_bulk_size;
- DMLC_DECLARE_PARAMETER(CachedOpParam) {
+ DMLC_DECLARE_PARAMETER(CachedOpConfig) {
DMLC_DECLARE_FIELD(inline_limit)
.set_default(2)
.describe("Maximum number of operators that can be inlined.");
@@ -96,8 +96,11 @@ class Imperative {
};
class CachedOp {
public:
- CachedOp(const nnvm::Symbol& sym,
- const std::vector<std::pair<std::string, std::string> >& kwargs);
+ CachedOp(
+ const nnvm::Symbol& sym,
+ const std::vector<std::pair<std::string, std::string> >& flags,
+ const std::vector<std::string> arg_names,
+ const std::unordered_map<std::string, std::vector<NDArray> >& params);
uint32_t num_inputs() {
return fwd_graph_.indexed_graph().input_nodes().size();
}
@@ -124,7 +127,7 @@ class Imperative {
std::vector<nnvm::NodeEntry> Gradient(const nnvm::NodePtr& node,
const std::vector<nnvm::NodeEntry>& ograds);
void Forward(const std::shared_ptr<CachedOp>& op_ptr,
- const std::vector<NDArray*>& inputs,
+ const std::vector<NDArray*>& args,
const std::vector<NDArray*>& outputs);
void Backward(const bool retain_graph,
const OpStatePtr& state,
@@ -138,14 +141,17 @@ class Imperative {
std::vector<OpStatePtr> states;
};
std::mutex mutex_;
- CachedOpParam param_;
+ CachedOpConfig config_;
nnvm::Graph fwd_graph_;
nnvm::Graph grad_graph_;
nnvm::Graph full_graph_;
+ std::unordered_map<Context, std::vector<NDArray> > params_;
bool inlining_;
std::vector<nnvm::NodeEntry> ograd_entries_;
std::vector<bool> curr_grad_req_;
std::vector<uint32_t> bwd_in_dep_, bwd_out_dep_, bwd_ograd_dep_;
+ std::vector<uint32_t> fwd_args_idx_;
+ std::vector<uint32_t> fwd_params_idx_;
std::vector<uint32_t> bwd_input_eid_;
std::vector<bool> save_inputs_, save_outputs_;
};
diff --git a/python/mxnet/_ctypes/ndarray.py b/python/mxnet/_ctypes/ndarray.py
index 20ad2bf..191985e 100644
--- a/python/mxnet/_ctypes/ndarray.py
+++ b/python/mxnet/_ctypes/ndarray.py
@@ -105,13 +105,28 @@ def _imperative_invoke(handle, ndargs, keys, vals, out):
class CachedOp(object):
"""Cached operator handle."""
__slots__ = ["handle"]
- def __init__(self, sym, flags=()):
+ def __init__(self, sym, flags=(), inputs=None, params=None):
self.handle = CachedOpHandle()
+ param_names = []
+ param_arrays = []
+ if inputs is None:
+ assert params is None, "When inputs is None params must also be None."
+ inputs = sym.list_inputs()
+ elif params is not None:
+ for name, arrs in params.items():
+ param_arrays.extend(arrs)
+ param_names.extend([name] * len(arrs))
+
check_call(_LIB.MXCreateCachedOpEx(
sym.handle,
len(flags),
c_str_array([key for key, _ in flags]),
c_str_array([str(val) for _, val in flags]),
+ len(inputs),
+ c_str_array(inputs),
+ len(params),
+ c_str_array(param_names),
+ c_handle_array(param_arrays),
ctypes.byref(self.handle)))
def __del__(self):
diff --git a/python/mxnet/gluon/block.py b/python/mxnet/gluon/block.py
index 0f41543..a737817 100644
--- a/python/mxnet/gluon/block.py
+++ b/python/mxnet/gluon/block.py
@@ -447,7 +447,6 @@ class HybridBlock(Block):
super(HybridBlock, self).__init__(prefix=prefix, params=params)
self._cached_graph = ()
self._cached_op = None
- self._cached_op_args = None
self._out_format = None
self._in_format = None
self._active = False
@@ -479,53 +478,54 @@ class HybridBlock(Block):
def _build_cache(self, *args):
inputs, out = self._get_graph(*args)
- input_idx = {var.name: i for i, var in enumerate(inputs)}
- self._cached_op = ndarray.CachedOp(out, self._flags)
- params = dict(self.collect_params().items())
-
- # verify graph inputs
- expected_inputs = set(out.list_inputs())
- for name in expected_inputs:
- assert name in params or name in input_idx, \
+ input_names = [i.name for i in inputs]
+
+ params = self.collect_params()
+ param_names = set(params.keys())
+ expected_names = set(out.list_inputs())
+ for name in expected_names:
+ assert name in param_names or name in input_names, \
"Unknown input to HybridBlock: %s"%name
- for name, i in input_idx.items():
- if name not in expected_inputs:
- warnings.warn("The %d-th input to HybridBlock is not used by any "
- "computation. Is this intended?"%i, stacklevel=4)
- for name in params:
- if name not in expected_inputs:
- warnings.warn("Parameter %s is not used by any computation. "
- "Is this intended?"%name, stacklevel=4)
-
- self._cached_op_args = [(False, params[name]) if name in params
- else (True, input_idx[name])
- for name in out.list_inputs()]
-
- def _finish_deferred_init(self, hybrid, *args):
+
+ used_input_names = [i for i in input_names if i in expected_names]
+ if len(used_input_names) != len(input_names):
+ unused = ', '.join(['%d-th'%i for i, name in enumerate(input_names)
+ if name not in expected_names])
+ warnings.warn("The %s input to HybridBlock is not used by any "
+ "computation. Is this intended?"%unused, stacklevel=4)
+
+ used_param_names = set(i for i in param_names if i in expected_names)
+ if len(used_param_names) != len(param_names):
+ unused = ', '.join(list(param_names - used_param_names))
+ warnings.warn("Parameter %s is not used by any computation. "
+ "Is this intended?"%unused, stacklevel=4)
+
+ used_params = {k: params[k] for k in used_param_names}
+ try:
+ param_dict = {k: v.list_data() for k, v in used_params.items()}
+ except DeferredInitializationError:
+ self._deferred_infer_shape(*args)
+ for i in used_params.values():
+ i._finish_deferred_init()
+ param_dict = {k: v.list_data() for k, v in used_params.items()}
+
+ self._cached_op = ndarray.CachedOp(out, self._flags, input_names, param_dict)
+
+ def _deferred_infer_shape(self, *args):
try:
self.infer_shape(*args)
except Exception as e:
error_msg = "Deferred initialization failed because shape"\
- " cannot be inferred \n {}".format(e)
+ " cannot be inferred. {}".format(e)
raise ValueError(error_msg)
- if hybrid:
- for is_arg, i in self._cached_op_args:
- if not is_arg:
- i._finish_deferred_init()
- else:
- for _, i in self.params.items():
- i._finish_deferred_init()
-
def _call_cached_op(self, *args):
if self._cached_op is None:
self._build_cache(*args)
args, fmt = _flatten(args, "input")
assert fmt == self._in_format, "Invalid input format"
- cargs = [args[i] if is_arg else i.data()
- for is_arg, i in self._cached_op_args]
- out = self._cached_op(*cargs)
+ out = self._cached_op(*args)
if isinstance(out, NDArray):
out = [out]
return _regroup(out, self._out_format)[0]
@@ -533,7 +533,6 @@ class HybridBlock(Block):
def _clear_cached_op(self):
self._cached_graph = ()
self._cached_op = None
- self._cached_op_args = None
def register_child(self, block, name=None):
if not isinstance(block, HybridBlock):
@@ -616,16 +615,17 @@ class HybridBlock(Block):
:py:class:`NDArray` or :py:class:`Symbol`."""
if isinstance(x, NDArray):
with x.context as ctx:
+ if self._active:
+ return self._call_cached_op(x, *args)
+
try:
- if self._active:
- return self._call_cached_op(x, *args)
params = {i: j.data(ctx) for i, j in self._reg_params.items()}
except DeferredInitializationError:
- self._finish_deferred_init(self._active, x, *args)
+ self._deferred_infer_shape(x, *args)
+ for _, i in self.params.items():
+ i._finish_deferred_init()
+ params = {i: j.data(ctx) for i, j in self._reg_params.items()}
- if self._active:
- return self._call_cached_op(x, *args)
- params = {i: j.data(ctx) for i, j in self._reg_params.items()}
return self.hybrid_forward(ndarray, x, *args, **params)
assert isinstance(x, Symbol), \
@@ -709,16 +709,10 @@ class SymbolBlock(HybridBlock):
self.params.get(i, grad_req='null', allow_deferred_init=True)
self._cached_graph = syms, out
- self._build_cache()
def forward(self, x, *args):
if isinstance(x, NDArray):
with x.context:
- try:
- return self._call_cached_op(x, *args)
- except DeferredInitializationError:
- self._finish_deferred_init(True, x, *args)
-
return self._call_cached_op(x, *args)
assert isinstance(x, Symbol), \
diff --git a/src/c_api/c_api_ndarray.cc b/src/c_api/c_api_ndarray.cc
index d67d52c..9aabe04 100644
--- a/src/c_api/c_api_ndarray.cc
+++ b/src/c_api/c_api_ndarray.cc
@@ -156,26 +156,47 @@ int MXCreateCachedOp(SymbolHandle handle,
nnvm::Symbol* sym = static_cast<nnvm::Symbol*>(handle);
API_BEGIN();
+ auto inputs = sym->ListInputs(nnvm::Symbol::kAll);
+ std::vector<std::string> input_names;
+ input_names.reserve(inputs.size());
+ for (const auto& i : inputs) input_names.push_back(i->attrs.name);
*out = new std::shared_ptr<Imperative::CachedOp>(
new Imperative::CachedOp(
- *sym, std::vector<std::pair<std::string, std::string> >()));
+ *sym,
+ std::vector<std::pair<std::string, std::string> >(),
+ input_names,
+ std::unordered_map<std::string, std::vector<NDArray> >()));
API_END();
}
int MXCreateCachedOpEx(SymbolHandle handle,
- int num_params,
+ int num_flags,
const char** keys,
const char** vals,
+ int num_args,
+ const char** arg_names,
+ int num_params,
+ const char** param_names,
+ NDArrayHandle* params,
CachedOpHandle *out) {
nnvm::Symbol* sym = static_cast<nnvm::Symbol*>(handle);
API_BEGIN();
- std::vector<std::pair<std::string, std::string> > kwargs;
+ std::vector<std::pair<std::string, std::string> > flags;
+ for (int i = 0; i < num_flags; ++i) {
+ flags.push_back({keys[i], vals[i]});
+ }
+ std::vector<std::string> args;
+ for (int i = 0; i < num_args; ++i) {
+ args.push_back(arg_names[i]);
+ }
+ std::unordered_map<std::string, std::vector<NDArray> > param_dict;
for (int i = 0; i < num_params; ++i) {
- kwargs.push_back({keys[i], vals[i]});
+ param_dict[param_names[i]].emplace_back(
+ *reinterpret_cast<NDArray*>(params[i]));
}
*out = new std::shared_ptr<Imperative::CachedOp>(
- new Imperative::CachedOp(*sym, kwargs));
+ new Imperative::CachedOp(*sym, flags, args, param_dict));
API_END();
}
diff --git a/src/executor/attach_op_execs_pass.cc b/src/executor/attach_op_execs_pass.cc
index e4d4955..b174ea0 100644
--- a/src/executor/attach_op_execs_pass.cc
+++ b/src/executor/attach_op_execs_pass.cc
@@ -241,8 +241,6 @@ Graph AttachOpExecs(Graph g) {
const auto& vdtype = g.GetAttr<DTypeVector>("dtype");
const auto& vshape = g.GetAttr<ShapeVector>("shape");
const auto& vctx = g.GetAttr<ContextVector>("context");
- const auto& saved_states = g.GetAttr<
- std::unordered_map<const nnvm::Node*, OpStatePtr> >("saved_states");
const auto& dispatch_modes = g.GetAttr<DispatchModeVector>("dispatch_mode");
// get the graph
@@ -271,13 +269,8 @@ Graph AttachOpExecs(Graph g) {
itype.emplace_back(vdtype[idx.entry_id(e)]);
}
- OpStatePtr state;
- if (saved_states.count(inode.source)) {
- state = saved_states.at(inode.source);
- } else {
- state = fcreate_op_state[op](
- inode.source->attrs, vctx[i], ishape, itype);
- }
+ OpStatePtr state = fcreate_op_state[op](
+ inode.source->attrs, vctx[i], ishape, itype);
FStatefulComputeEx fcompute_ex = common::GetFCompute<FStatefulComputeEx>(
op, "FStatefulComputeEx", vctx[i]);
// FStatefulComputeEx is dispatched only when dispatch_mode is DispatchMode::kFComputeEx
diff --git a/src/executor/graph_executor.cc b/src/executor/graph_executor.cc
index 4d24f55..d5dacf7 100644
--- a/src/executor/graph_executor.cc
+++ b/src/executor/graph_executor.cc
@@ -904,7 +904,6 @@ void GraphExecutor::FinishInitGraph(nnvm::Symbol symbol,
}
g = DetectInplaceAddTo(g);
- g.attrs["saved_states"] = std::make_shared<nnvm::any>(std::move(saved_states_));
g = AttachOpExecs(g);
g = AttachOpResources(g);
graph_ = std::move(g);
diff --git a/src/executor/graph_executor.h b/src/executor/graph_executor.h
index 3f1ebe5..bcde41d 100644
--- a/src/executor/graph_executor.h
+++ b/src/executor/graph_executor.h
@@ -234,8 +234,6 @@ class GraphExecutor : public Executor {
size_t num_forward_inputs_{0};
// number of forward nodes
size_t num_forward_nodes_{0};
- // saved operator for autograd
- std::unordered_map<const nnvm::Node*, OpStatePtr> saved_states_;
// monitor call back
std::function<void(const char*, void*)> monitor_callback_{nullptr};
// whether to enable bulk execution
diff --git a/src/imperative/cached_op.cc b/src/imperative/cached_op.cc
index ce23735..10a8fc0 100644
--- a/src/imperative/cached_op.cc
+++ b/src/imperative/cached_op.cc
@@ -22,17 +22,19 @@
namespace mxnet {
-DMLC_REGISTER_PARAMETER(CachedOpParam);
+DMLC_REGISTER_PARAMETER(CachedOpConfig);
Imperative::CachedOp::CachedOp(
const nnvm::Symbol& sym,
- const std::vector<std::pair<std::string, std::string> >& kwargs) {
+ const std::vector<std::pair<std::string, std::string> >& flags,
+ const std::vector<std::string> arg_names,
+ const std::unordered_map<std::string, std::vector<NDArray> >& params) {
using namespace nnvm;
using namespace imperative;
static const std::vector<const Op*> zero_ops{Op::Get("zeros_like"), Op::Get("_zeros")};
static const auto _copy = Op::Get("_copy");
- param_.Init(kwargs);
+ config_.Init(flags);
// construct forward graph
{
@@ -66,7 +68,34 @@ Imperative::CachedOp::CachedOp(
fwd_graph_.attrs["forward_ref_count"] =
std::make_shared<dmlc::any>(std::move(ref_count));
- inlining_ = (idx.num_nodes() - idx.input_nodes().size()) <= param_.inline_limit;
+ inlining_ = (idx.num_nodes() - idx.input_nodes().size()) <= config_.inline_limit;
+ }
+
+ // Set params
+ {
+ const auto& idx = fwd_graph_.indexed_graph();
+ std::unordered_map<std::string, size_t> arg_name_to_id;
+ for (size_t i = 0; i < idx.input_nodes().size(); ++i) {
+ const auto& name = idx[idx.input_nodes()[i]].source->attrs.name;
+ auto iter = params.find(name);
+ if (iter == params.end()) {
+ arg_name_to_id[name] = i;
+ continue;
+ }
+ fwd_params_idx_.push_back(i);
+ for (const auto& param : iter->second) {
+ params_[param.ctx()].emplace_back(param);
+ }
+ }
+
+ CHECK_EQ(arg_name_to_id.size(), arg_names.size())
+ << "Expecting " << arg_name_to_id.size() << "inputs, given " << arg_names.size();
+
+ for (const auto& name : arg_names) {
+ auto iter = arg_name_to_id.find(name);
+ CHECK(iter != arg_name_to_id.end()) << "Unexpected input name " << name;
+ fwd_args_idx_.push_back(iter->second);
+ }
}
// construct backward graph
@@ -341,22 +370,38 @@ nnvm::Graph Imperative::CachedOp::GetBackwardGraph(
void Imperative::CachedOp::Forward(
const std::shared_ptr<CachedOp>& op_ptr,
- const std::vector<NDArray*>& inputs,
+ const std::vector<NDArray*>& args,
const std::vector<NDArray*>& outputs) {
using namespace nnvm;
using namespace imperative;
static const auto cached_op = nnvm::Op::Get("_CachedOp");
+ CHECK_EQ(args.size(), fwd_args_idx_.size())
+ << "CachedOp requires " << fwd_args_idx_.size()
+ << " inputs but got " << args.size();
+
+ Context default_ctx = args[0]->ctx();
+
+
+ std::vector<NDArray*> inputs(num_inputs());
+ for (index_t i = 0; i < fwd_args_idx_.size(); ++i) {
+ inputs[fwd_args_idx_[i]] = args[i];
+ }
+ if (fwd_params_idx_.size()) {
+ CHECK(params_.find(default_ctx) != params_.end())
+ << "CachedOp is not initialized on context " << default_ctx;
+
+ for (size_t i = 0; i < fwd_params_idx_.size(); ++i) {
+ inputs[fwd_params_idx_[i]] = ¶ms_[default_ctx][i];
+ }
+ }
+
// Initialize
bool recording = Imperative::Get()->is_recording();
nnvm::Graph g = GetForwardGraph(recording, inputs);
const auto& idx = g.indexed_graph();
size_t num_inputs = idx.input_nodes().size();
- CHECK_EQ(num_inputs, inputs.size())
- << "CachedOp requires " << num_inputs << " but got " << inputs.size();
-
- Context default_ctx = inputs[0]->ctx();
for (size_t i = 0; i < inputs.size(); ++i) {
CHECK_EQ(inputs[i]->ctx(), default_ctx)
<< "CachedOp requires all inputs to live on the same context. But "
@@ -403,7 +448,7 @@ void Imperative::CachedOp::Forward(
const auto& dispatch_modes = g.GetAttr<DispatchModeVector>("dispatch_mode");
if (recording && !inlining_) Imperative::Get()->set_is_recording(false);
- int prev_bulk_size = Engine::Get()->set_bulk_size(param_.forward_bulk_size);
+ int prev_bulk_size = Engine::Get()->set_bulk_size(config_.forward_bulk_size);
Imperative::Get()->RunGraph(
false, idx, arrays, 0, idx.num_nodes(), std::move(array_reqs),
@@ -491,7 +536,7 @@ void Imperative::CachedOp::Backward(
const auto& dispatch_modes = g.GetAttr<DispatchModeVector>("dispatch_mode");
- int prev_bulk_size = Engine::Get()->set_bulk_size(param_.backward_bulk_size);
+ int prev_bulk_size = Engine::Get()->set_bulk_size(config_.backward_bulk_size);
Imperative::Get()->RunGraph(
retain_graph, idx, arrays, num_forward_nodes, idx.num_nodes(),
--
To stop receiving notification emails like this one, please contact
jxie@apache.org.