Summary:
As in the title. Lets us simplify a lot of code.
Depends on #9363, so please review only the last commit.
zdevito
Pull Request resolved: https://github.com/pytorch/pytorch/pull/9414
Reviewed By: zdevito
Differential Revision: D8836496
Pulled By: apaszke
fbshipit-source-id: 9b3c3d1f001a9dc522f8478abc005b6b86cfa3e3
* pad-sequence no longer requires sorting entries
pad-sequence can get the max_len from the list of sequences. entries only need to be sorted if output will be used for pack_padded_sequence, which can throw the error itself.
* remove sort requirement from pad-sequence
Picks up from #5974.
Removes the requirement that input sequences to pad_sequence have to be
sorted. Addressed the comments in the PR:
- Updated docstring for pad_sequence
- Remove sort requirement in pad_sequence test
- Test unsorted and sorted sequences in pad_sequence test
* Fix handling of empty batches in SumReduceDimsOp
As titled
* Deferrable async_scheduling finishRun fix
Proper order of finishing run operations in deferrable_async_scheduling net
* Simplify exception handling in async_scheduling
Simplify exception handling, no need to busy wait, thread that processes the
last task can finish the run
* [C2]worker_coordinator_memorize_worker_ids
As titled. This is related to T28689868, where the number of blobs we want to create is equal to the number of worker ids
* Add unit test for nets with no type set
* Ignore total length argument in sympolic_pad_packed_sequence
1- There was a mistake in the code that total_length was added to the wrong symbolic function (pack_padded_sequence) instead of (pad_packed_sequence)
2- No need to throw an exception if total_length is given since it is only used to enable data_parallel training on multi-gpus and doesn't have anything to do with onnx export, so just ignore it. https://fburl.com/tk4gciqp
* Add support for MKLDNN to async_scheduling
Just add MKLDNN as a possible CPU option to async_scheduling's pool function
* [AuFL][ensemble] support branch output for prediction
This diff supports using predictions from different branches and thus enables model ensembling (not fully independent).
* Fix a bug in add_loss in layer_model_helper
As titled.
* Support lradaption for adam
1.lr adaption operator
2.apply to dense adam
* Perf tweaks for async_scheduling
Restore single pool option + remove unnecessary (no-ops) calls
* add quantization to SparseSimdAdagradOp
add a bunch of quantization signatures to SparseSimdAdagradOp, implementations to come next
* [sr] [codemod] Change all SR callsites to use new API
@allow-large-files
This diff refactors all callsites of SR to use the slightly changed API introduced in the diff below. Really what this means is that you need to include the correct header. Also if you were using `ClientFactory::newFactory` you need to not prefix it with `ClientFactory::`.
```
cd ~/fbsource/fbcode
find ./ -type f -exec sed -i -e 's:#include "servicerouter/client/cpp2/ClientFactory.h":#include "servicerouter/client/cpp2/ServiceRouter.h":' -e 's:#include <servicerouter/client/cpp2/ClientFactory.h>:#include <servicerouter/client/cpp2/ServiceRouter.h>:' -e 's/ClientFactory::newFactory(/newFactory(/g' {} \;
```
Also manually fixed spots that couldn't be done automatically (or broke because they depended on transitive includes).
* Back out "Fix handling of empty batches in SumReduceDimsOp"
Original commit changeset: 282da1730cc2 This commit is blocking the
Github->fbcode sync, which really needs to get merged ASAP. D7881937 which this
diff depends on will be reverted in the sync D7990948 which causes this to
break. The sync diff cannot be patched with this reversion because it must be
landed against base revision 5c8c099 , and D7881937 must not be included in the
sync diff because it is breaking GPU tests that are not available in sandcastle
: https://ci.pytorch.org/jenkins/job/caffe2-builds/job/py2-cuda8.0-cudnn6-ubuntu16.04-test/3638/console
for one example.
* Add the flow to support operator benchmark
1) generate model with the operator 2) upload to everstore 3) generate model spec into json file 4) start running the benchmark
* [tum][gpu] Connect DPM trainer with flow and unit tests
This diff:
- Fix some small bugs for Yiming's recent changes to parallelizer, so it suits real use cases.
- Add correct tags to the TUM code, so we can do data parallel transform
- pass extra info when instantiation.
- add unit test for using DPM in TUM model
After this diff, we can do simple box, multi-gpu fully-sync trainer for TUM in Fblearner workflow, but may still need to do speed benchmarking.
* w/o normalized lradaption for adam dense only
The previous lr adaption includes a normalization step when performing the dot product operation. This is not exactly same as what is proposed in the paper. I add normalization as an option. Without it, the operator performs exactly what the paper proposed. With the option, we add the normalization step
* [fb] Use SharedPromise in DeferrableAsyncSchedulingNet
This code is to simplify DeferrableAsyncSchedulingNet by removing condition
variable + small fixes
* [tum] implement cuda sparseLengthsMean and LengthsMean
as title
* Adding an optional parameter to allow use of protobufs in InferShapesAndTypes function.
Adding an optional parameter to allow use of protobufs in InferShapesAndTypes function.
* Move feature_to_index to FeatureSpec.feature_to_index
move feature_to_index to FeatureSpec.feature_to_index to avoid override other fields
* [Caffe2] Rename bytes_moved to bytes_written
Just a rename in preparation for supporting bytes_read.
* [c2] fix ReduceFrontSumOp for empty case by setting 0
otherwise, it may use the results from last iteration when it's empty batch.
* [Caffe2] [Int8] Improve Intel CPU performance
* [Easy] Improve PrependDim op logging
as titled
* DBFileReader expand db_path using os.path.expanduser(..)
Since there are a lot of possible use cases of `DBFileReader` to read from user home path, like `~/local/sample.db`, I want to save people's trouble of calling `os.path.expanduser(db_path)` themselves.
* [Caffe2] Add bytes_read to cost structure
We're adding analytical read bytes to cost functions. This extends the structure accordingly for all CostInference defined operators.
Additionally, some small bug fixes were performed:
1) Cost functions now extract type information of operands instead of assuming float
* Fix sleef on aarch64 for hhvm
@bypass-lint
Rename flag
* Remove duplicated part in caffe2/ideep/operators/conv_op.cc
should be sync error
* Rename test helper function test_adagrad_sparse_helper to adagrad_sparse_test_helper to avoid confusing pytest
* ENH: add to method for PackedSequence
* ENH: return self if possible
* TST: remove extra data
* DOC: add more explanation
* TST: remove extra data
* DOC: minor fix
* Codemod to update our codebase to 0.4 standard
* Update some of the test scri[ts
* remove Variable in test_clip_grad_value
* fix _symbolic_override_wrapper_maker
* Unit test for pack_padded tracing
* Move monkeypatching stuff
* Switch symbolic
* Fix stack traces and update test
* Fixup and confirm e2e working
* lint
* Move monkeypatch back to onnx
* Address comments
* remove extraneous import
* Add gradient checking
* lint
* Address comments
* improve test case
* Namespaced symbols
- Our interned strings now have structure, "ns::symname" rather than just
"symname" before. We support efficient namespace testing for uniques
by encoding the namespace in one byte in the Symbol internal representation.
See torch/csrc/jit/interned_strings.h for a more in-depth implementation
discussion.
- All uses of ksymbol are now attr::symbol (or some appropriate namespace).
The valid namespaces are prim, attr, onnx and aten.
- Symbol is bound in Python as a qualified string "attr::symbol", EXCEPT for the
attribute setting/getting API, whose symbols must always be attr
symbols; they get special cased to assume strings are passed.
There's a little bit of naughtiness in the implementation, maybe you know
how to solve it.
- However, the g.op() convenience function assumes that you're generating
ONNX operators, unless you explicitly qualify.
- All ATen operators and nodes have built-in interned strings generated
for them, so you should never have to write a string literal ever again.
The tracing code is adjusted to use it.
- ONNX exporter now properly tests to see that all operators are in
onnx namespace before accepting the export. This is way more
robust than the previous exporter, which would be willing to
export capitalized operators which were not actually ONNX operators.
- A slight organizational change for symbolic.py; this module now ONLY
contains aten operators. In particular, the exporter for Constant
has moved into utils.py (along with Undefined, from the C++ side),
since primitive ops get "special treatment."
- The un-inplacing logic in recording is more robust, so that we don't
delete a trailing underscore from __and__. This never affected us
before because we didn't have any tests for it.
Signed-off-by: Edward Z. Yang <ezyang@fb.com>
* Fix some minor errors in existing docs.
* Fix Convolution and Pooling docs in torch.nn.functional
* Cleaned up torch.nn.functional docs
* Address @SsnL 's comments
* Add multiplication sign missing in docs
* Fix more typos, and clear some warnings
* Change infinity symbol in LPPool2d
* Revert some changes in torch.nn.functional
* Few more minor changes
* Improvize documentation
1. Add formula for erf, erfinv
2. Make exp, expm1 similar to log, log1p
3. Symbol change in ge, le, ne, isnan
* Fix minor nit in the docstring
* More doc improvements
1. Added some formulae
2. Complete scanning till "Other Operations" in Tensor docs
* Add more changes
1. Modify all torch.Tensor wherever required
* Fix Conv docs
1. Fix minor nits in the references for LAPACK routines
* Improve Pooling docs
1. Fix lint error
* Improve docs for RNN, Normalization and Padding
1. Fix flake8 error for pooling
* Final fixes for torch.nn.* docs.
1. Improve Loss Function documentation
2. Improve Vision Layers documentation
* Fix lint error
* Improve docstrings in torch.nn.init
* Fix lint error
* Fix minor error in torch.nn.init.sparse
* Fix Activation and Utils Docs
1. Fix Math Errors
2. Add explicit clean to Makefile in docs to prevent running graph generation script
while cleaning
3. Fix utils docs
* Make PYCMD a Makefile argument, clear up prints in the build_activation_images.py
* Fix batch norm doc error
This was accidentally lost while addressing review comments on
https://github.com/pytorch/pytorch/pull/4695
pack_padded_sequence may be called either with a list or with a
Variable. If called with a list we convert to Variable internally.
I added to test_nn to test the new codepath. The bug was also caught
by the onnx-fb-universe tests (which rely on passing in Variable).
* PackedSequence: store batch_sizes as tensor
rather than converting to a list of python integers. This maintains
the invariant that module's inputs/outputs are collections of
Variables.
In particular, this causes the JIT to no longer choke when flattening
and unflattening arguments.
* Handle sequence lengths correctly when exporting RNNs to ONNX
- when uniform sequence lengths are provided, correctly omit the
argument when constructing the ONNX graph, so as to not fix the
graph to the batch size.
- handle PackedSequences by floating them through the graph and
eliminating them in an optimization pass. ONNX does not have packed
sequences, but operates on a representation equivalent to
PaddedSequence, so we hide the representation-switching from ONNX
- as a preliminary step towards handling PackedSequences, not directly
tied to ONNX export, change batch_sizes from being an argument to
the RNN operators into being an argument to the forward() function
of those RNN operators. This more closely models the reality that
batch_sizes are effectively part of the input sequences.
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.
* Specifying the value used for padding
The "pad_packed_sequence" function fills padded elements with zeros, but sometimes it is not useful. For example, some previous papers on NLP, including my recent paper [1], use a max-pooling technique for RNN-based sentence representations. More specifically, the max-pooling technique selects the maximum value from all time steps (i.e., hidden states) for each dimension. In such a case, we do not want the padded zeros to be selected. To overcome this situation, we can simply use a very small value instead of zero.
An LSTM example is shown below:
input = embedding(Variable(batchInput))
packedInput = nn.utils.rnn.pack_padded_sequence(input, lengths, batch_first = True)
h, (hn, cn) = self.encoder(packedInput, (h0, c0))
h, _ = nn.utils.rnn.pad_packed_sequence(h, -1024.0 batch_first = True)
sentenceRep, _ = torch.max(h, 1, keepdim = True)
[1] A Joint Many-Task Model: Growing a Neural Network for Multiple NLP Tasks. Kazuma Hashimoto, Caiming Xiong, Yoshimasa Tsuruoka, and Richard Socher. The 2017 Conference on Empirical Methods in Natural Language Processing (EMNLP 2017).
https://arxiv.org/abs/1611.01587 (Equation (4))
* Modified the order of the arguments
Following the suggestion, I modified the order of the arguments.
