This replaces the torch.Tensor constructors with factories that produce
Variables. Similarly, functions on the torch module (e.g. torch.randn)
now return Variables.
To keep the PR to a reasonable size, I've left most of the unused tensor
code. Subsequent PRs will remove the dead code, clean-up calls to
torch.autograd.Variable, and rename Variable to Tensor everywhere.
There are some breaking changes because Variable and Tensors had
slightly different semantics. There's a list of those changes here:
https://github.com/pytorch/pytorch/wiki/Breaking-Changes-from-Variable-and-Tensor-merge
* Improve Function interface
* Undo tracer changes
* Fix bug in VariableType.set_history
* Rename function_counter and sequence_number to sequence_nr
* Clarify Function documentation
* Replace swap_next_edges with next_edges() getter
* Bring back set_gradient_edge
* Simplify special.cpp
* add_gradient_edge -> create_gradient_edge
* Add mutable getters for pre/post hooks
* Use make_variable with Edge
* Remove remove_gradient_edge in favor of detach_
* Fix documentation and remove create_gradient_edge friend method
* Canonicalize some includes
* Add numpy-style dtypes to Variable factories.
1) Add numpy-style dtypes corresponding to torch tensor types. These are:
torch.float16, torch.float32, torch.float64, torch.uint8, torch.int8, torch.int16, torch.int32, torch.int64
as well as torch.cuda, torch.sparse, and torch.cuda.sparse equivalents.
2) Adds "legacy" names for the above dtypes that correspond more closely to existing tensor names. These are:
torch.half, torch.float, torch.double, torch.short, torch.int, torch.long.
torch.byte and torch.char don't exist because they either don't match numpy semantics or differ on different architectures.
3) Adds a "dtype" parameter to Variable factories (e.g. zeros, ones) that allows the user to specify the type without changing the default tensor type.
4) Adds a "dtype" getter to Variables that return the canonical dtype from 1)
This PR is missing the following useful features that should be added in the future:
A) We only add the "dtype" parameter to auto-generated factories; hand-written factories like in tensor_new.cpp don't support this yet.
B) We don't allow type conversions to use dtypes; that should be added to type(param) or a new function.
C) We don't yet have a "device" parameter for these factories; right now, they will only create Variables on the default device.
* backend_to_string can be private.
* Define python binding argument indexes in a more simple way.
* add all_declared_types, still need to hook it up to THPDType.
* Fix all_declared_types for missing types (it's Sparse + Half).
* Ensure cuda dtypes are created even if compiled with NO_CUDA=1.
* Fix case where dtype is provided but dispatch is via namespace.
This happens in ones_like, empty_like, randn_like.
There is some question if we should do:
1) at::ones_like(tensor).toType(dtype)
2) at::ones_like(tensor.toType(dtype))
I did the former because this matches with the numpy documentation, i.e.:
"Overrides the data type of the result." and it's easier to implement.
Note that the above causes an extra copy, either of the input or output.
Here's a better implementation:
1) Make zeros_like, ones_like native functions that take an optional type (named dtype?).
2) Match the type argument with the dtype, so we don't have two different parameters.
3) Call at::zeros_like(input, type) -> at::native::zeros_like(input, type) -> type.zeros(input.sizes())
* Don't return from maybe_initialize_cuda.
* Don't leak DType name.
* Address cpp review comments.
* Share code between sparse and non-sparse test_dtypes.
* Rewrite _like functions as native function with explicit type parameter.
* Use type 'Type' instead of 'dtype' for consistency.
* Address review comments.
* Handle arg_idx when there is requires_grad but no dtype in python_binding_arguments.
* Improve Variable interface
* Address comments from @apaszke and @colesbury
* string ::operator= is not noexcept
* Remove ir.h from tracer_state.h to improve build times
* Make Variable a struct and pack SavedVariable fields
* Implement as_variable_ref
* grad_fn_ptr() -> grad_fn_unsafe()
* Reduce hackiness of set_type hack
* Include variable.h and edge.h in tracer_state.h because it uses them
* class Variable -> struct Variable because Windows cant even
* Make Variable::output_nr uint32_t instead of int
* Add comment about tracing state
* Replaced more static_cast<Variable&> and improve docs
* Remove SavedVariable destructor and construct members in init list
* Clarify docs for Variable
* Variable::set_version -> set_version_counter
Don't check the ScalarType and Backend of arguments in VariableType.
Instead, only check that arguments are Variables of any type. The
precise type checks are handled by the base type.
Many of our functions take heterogeneous types. There isn't enough
information in Declarations.yaml to ensure the precise types of
arguments in VariableType, which makes it difficult to add new methods.
This is #4943 with a fix to the memset call
Don't check the ScalarType and Backend of arguments in VariableType.
Instead, only check that arguments are Variables of any type. The
precise type checks are handled by the base type.
Many of our functions take heterogeneous types. There isn't enough
information in Declarations.yaml to ensure the precise types of
arguments in VariableType, which makes it difficult to add new methods.
* Enable scalars if compiled with WITH_SCALAR environment variable.
We are pretty close to enabling scalars (0-dimensional arrays); this allows turning them on
for development purposes and to be able to write code that works both with and without scalars enabled.
WITH_SCALARS is currently broken with distributions, but should work for test_torch, test_autograd, test_nn.
* Fix unsqueeze.
* Fix wrap dim, wrapping with Scalar.
This adds overrides in VariableType for the xxx_out ATen functions and
implements Python bindings. There is no support for automatic
differentiation. If any of the inputs (or outputs) requires grad, then the
function will throw an exception unless it's running in "no-grad" mode.
The bindings for calling torch.xxx functions on Variables are moved to a
different object. Previously, they were static method on VariableBase.
This change prevents users from accidentally calling static methods as if
they were instance methods.
This removes volatile from Variable. The functionality is mostly
replaced by a global (thread-local) flag, which is controlled by
torch.set_grad_enabled() and the context manager torch.no_grad().
In C++, the flag is exposed through GradMode::is_enabled() and GradMode::set_enabled()
Fixes#3627
Implements basic and advanced indexing using ATen tensors/variables.
Basic indexing is translated at the Python-binding level
(python_variable_indexing.cpp) to slice/squeeze/unsqueeze/select calls.
Advanced indexing is implemented in ATen in terms of take() and put()
calls.
* Use aten version of is_signed.
* Define is_cuda native function and use it for variable.
* Use ATen dim for Variable dim/ndimension.
* Get rid of dim, ndimension fallthroughs in variable.py.
* Move size/stride Variable methods to use ATen.
* Implement shape property on Variable via ATen.
* Remove the _getattr__ function from Variable.
* Get rid of dispatch functions and avoid cast.
* Add THPUtils_packInt64Array.
* Throw python errors.
* Use fallthrough and fix fallthrough generation for native functions.
* is_cuda is a property, not a method.
Add (fully opt-in) functionality to support setting pretty names for
nodes in the graph. In particular
- Variable now has a `name` parameter in the constructor
- export now has `input_names` and `export_names` parameters
Nodes that are not named via this mechanism continue to be named
internally with unique integers.
Names have a few rules.
- They must all be unique in the graph.
- They may not be integers (because of potential conflicts with
internally generated names).
Allow in-place operations on views
Adds VariableViewImpl, a subclass of VariableImpl which has a pointer to
the base Variable on which it is a view. In-place operations on views
change the grad_fn of the base.
Note that in-place operations only work on views that are the first output of the function that created them. All C++/ATen implemented functions have this behavior, but it's possible to write Python-implemented autograd functions that do not. In-place operations on these view will raise an exception.
Fixes#3313
Previously, we the Variable.data PyObject* in THPVariable_Wrap. For many
Variables, we don't access their data directly. Instead, they are passed
from one Variable compuatation to another.
This reduces the overhead of ATen-implemented Variable methods by
~200ns.
Also squash a warning about an implicit conversion that will never
occur (because the type being converted to is a superclass).
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
Variable is now a subclass of at::Tensor backed by a VariableImpl* pImpl. The implementation of the ATen functions is defined in the auto-generated VariableType.h/cpp file.
Currently, only functions which fall through to the base type, such as sizes() and isCuda() are implemented. Differentiable ops like add() and mul() will be added in a subsequent PR.
Previously, our AST was a DAG, where shared Nodes indicated a computation
should be reused. This commit rewrites the IR into a new functional
representation which represents sharing explicitly using variable
bindings.
We offer a few justifications for this new style:
1. The new representation is not all that different from the
old one; it is about as easy to construct, and the lack of an
explicit graph doesn't negatively impact our ability to interpret
the graph, since we've chosen, as a matter of design, to NOT have
the IR participate in the actual execution of a graph.
2. The new let-binding representation has an implicit ordering,
which we can use to conveniently keep track of the original order
the trace showed up as. This automatically gives us a topsort,
and gives us an easier to read textual representation of our
IR:
%14 = Embedding %11, %0, -1, None, 2, False, False
%15 = Dropout %14, 0.2, True, False
%16 = Index %12, 0
%17 = Index %12, 1
%18 = Index %13, 0
%19 = Index %13, 1
%20 = Index %15, 0
%21 = Linear %20, %1, %3
%22 = Linear %16, %2, %4
3. It moves us closer to a Futhark style language
(http://futhark-lang.org/publications/pldi17.pdf).
Major aspects of the diff
- Node is replaced with Expr and Arg, a pair of mutually recursive
structures which represent our new language. In BNF, the language
looks like this:
a ::= c | %i
e ::= %i, ... = e
| PyOp e, ...
| Ret %i, ...
Technically, Ret is not actually a return (no control flow is involved),
it just tuples up a series of tensors (identified by variables).
One important invariant is that locals are always tensors; they
are never constants (this is asymmetric with Args.)
- Arguments support Python constants. This is an important piece because
many operators take extra Python literals like integers and tuples in
order to specify extra parameters about how an operator operates. Adding
this was essential to getting word_language_model to work.
- As both Expr and Arg have multiple variants, there is new infrastructure
for doing case on the variants using ExprVisitor and ArgVisitor. The
strategy here is adapted from WebAssembly's visitors, although we have
generalized to permit arbitrary argument forwarding, which is necessary
to support tail-recursive visitor calls. TCO is important because our
interpreter may recurse arbitrarily deep into a stack of nested lets.
If users wish, they can also manually case on the type tag.
- Tracing is now turned on and off using _tracer_enter/_tracer_exit in
torch._C. _tracer_enter accepts a list of variables which are to be
treated as arguments; _tracer_exit accepts the list of traced variables
which should be returned when you reexecute the trace, and returns
the trace expression which can be reexecuted. GlobalTracingState
is a global variable which tracks whether or not we are tracing or not.
- You use run_forward to execute a trace on some set of parameters.
- When under tracing, variables keep track, via trace_local, what the
name of their variables in the IR are.
Here is a simple runner which leaks memory but can be used to JIT models:
import torch.autograd.function as F
import torch._C
def jit(model):
import types
real_forward = model.forward
def forward(self, *args):
def flatten(x):
return tuple(F._iter_variables(x))
if not hasattr(self, "saved_trace"):
torch._C._tracer_enter(tuple(self.parameters()) + flatten(args))
out = real_forward(*args)
self.saved_trace = torch._C._tracer_exit(flatten(out))
self.saved_outs = out
return out
else:
flat_out = Variable._execution_engine.run_forward(self.saved_trace, tuple(self.parameters()) + flatten(args))
return F._unflatten(flat_out, self.saved_outs)
Major problems:
- Sanity checking is spotty at best, especially when users pass in variables.
- The interpreter leaks tensor memory from the store. When we add back def-use
we should be able to deallocate tensors as soon as we know they are no longer
necessary.
- The interpreter needs to reach feature parity with the old execution engine.
From there, we need to see if backwards can be subsumed as well.
- I still have no confidence in having memory managed everything correctly.
This requires a close look.
- Rather than return an *open* expression as a trace, we should return a
*lambda* instead, which knows about how many formal parameters it
requires.
- The IR is not introspectable from Python at the moment, but this is simply a
matter of implementing all the binding code.
- The tracer is NOT reentrant (you can't trace while you're inside a trace.)
Furthermore, no sanity checking is done if you try to incorrectly reuse
things from one trace in another.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
* Fix gc_refs assertion failure
Ensure that each THPVariable -> THPFunction reference contributes one
ref count to the THPFunction by creating a new shared_ptr for each ref.
Because multiple shared_ptrs can again manage a single THPFunction, it's
not safe to use std::weak_ptr where it may point to a PyFunction. It's
still safe to use weak_ptr for grad_accumulator since these are never
PyFunctions.
Fixes#1626
* Remove stale comment
Because of this Variables can no longer appear in the graph.
Every usage of a leaf Variable will leave an AccumulateGrad
function that has no outputs, but modifies var.grad as a side
effect.
The core autograd Variable, Function, and Engine no longer depend on the
Python API. This let's us implement functions in C++. In the future, we
can also multithread engine and release the GIL for most of the
non-Python backwards.
