Summary: As pointed out in https://github.com/pytorch/pytorch/pull/107479, using a set prevents collisions like "a" => "a", "a" => "a_1", "a_1" => "a_1" (but should go to "a_1_1"). We can combine using counters and a set to avoid this problem. Still gets us the performance benefit in the case of collisions with a very minor penalty in a case with no collision.
Test Plan:
Extract this code and run:
```
# New version
from typing import Dict, Set
class Net:
_net_names_used_counters: Dict[str, int] = {}
_net_names_used: Set[str] = set()
staticmethod
def current_prefix():
return "test_prefix"
staticmethod
def _get_next_net_name(basename):
basename = "/".join(x for x in [Net.current_prefix(), basename] if x)
idx = Net._net_names_used_counters.get(basename, 0)
while (name := basename if idx == 0 else f"{basename}_{idx}") in Net._net_names_used:
idx += 1
Net._net_names_used_counters[basename] = idx + 1
Net._net_names_used.add(name)
return name
print(Net._get_next_net_name("basename"))
print(Net._get_next_net_name("x_basename"))
print(Net._get_next_net_name("basename"))
print(Net._get_next_net_name("basename"))
print(Net._get_next_net_name("x_basename"))
print(Net._get_next_net_name("basename_1"))
> test_prefix/basename
> test_prefix/x_basename
> test_prefix/basename_1
> test_prefix/basename_2
> test_prefix/x_basename_1
> test_prefix/basename_1_1
```
Differential Revision: D48576516
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107743
Approved by: https://github.com/zdevito
Summary:
There is a module called `2to3` which you can target for future specifically to remove these, the directory of `caffe2` has the most redundant imports:
```2to3 -f future -w caffe2```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/45033
Reviewed By: seemethere
Differential Revision: D23808648
Pulled By: bugra
fbshipit-source-id: 38971900f0fe43ab44a9168e57f2307580d36a38
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/42927
added fp16 fusion to net transforms
refactored the transforms as well as glow_transform to get out of opt/custom so that the OSS builds passed
Test Plan: added net runner tests for this
Reviewed By: yinghai
Differential Revision: D23080881
fbshipit-source-id: ee6451811fedfd07c6560c178229854bca29301f
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/42763
add the fp16 fusions as net transforms:
-layernorm fused with mul+add
-swish int8
Test Plan: added unit test, ran flows
Reviewed By: yinghai
Differential Revision: D23002043
fbshipit-source-id: f0b13d51d68c240b05d2a237a7fb8273e996328b
Summary:
Changes in PR https://github.com/pytorch/pytorch/issues/39759 broke HIP caffe2.
hipify for caffe2 renames CUDA to HIP; torch does not.
If caffe2 calls into torch, it needs to use CUDA-named functions.
CC ezyang xw285cornell sunway513 houseroad dzhulgakov
Pull Request resolved: https://github.com/pytorch/pytorch/pull/39801
Differential Revision: D21982493
Pulled By: xw285cornell
fbshipit-source-id: 8e88e0fb80c71f0342e23ef0214a42d5542bdc70
Summary:
THCAllocator functionality is pretty obscure and it's hard to get it working with HIP because of how Caffe2/PyTorch rules are set up (see https://github.com/pytorch/pytorch/issues/39801). Let's just disable the test.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/39843
Reviewed By: zou3519
Differential Revision: D21998687
Pulled By: dzhulgakov
fbshipit-source-id: cd12ba30cdfee658b98393ed3a72e83f4ecf1c9c
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/39759
Caffe2 has a mode where it uses PT's caching allocator. Somehow we were not calling the initialization explicitly.
Now, I have no idea why it worked before. Probably worth to run a bisect separately.
Reviewed By: houseroad
Differential Revision: D21962331
fbshipit-source-id: f16ad6b27a67dbe0bda93939cca8c94620d22a09
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/18560
We have to import python protobuf here **before** we load cpp extension.
Otherwise it breaks under certain build conditions if cpp implementation of
protobuf is used. Presumably there's some registry in protobuf library and
python side has to initialize the dictionary first, before static
initialization in python extension does so. Otherwise, duplicated protobuf
descriptors will be created and it can lead to obscure errors like
Parameter to MergeFrom() must be instance of same class: expected caffe2.NetDef got caffe2.NetDef.
I think it also fixes https://github.com/facebookarchive/caffe2/issues/1573
Reviewed By: ezyang, iroot900
Differential Revision: D14622054
fbshipit-source-id: 2499eb88ecdee85ff8d845859048f7ae5da2a480
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/17726
Pull Request resolved: https://github.com/pytorch/pytorch/pull/17725
Pull Request resolved: https://github.com/pytorch/pytorch/pull/17461
Implementing a standalone LSTM Operator in Caffe2 adopted from this Aten implementation: diffusion/FBS/browse/master/fbcode/caffe2/aten/src/ATen/native/RNN.cpp. The most tricky thing in this exercise was that caffe2::Tensor has no copy constructor that made it necessary to implement a custom templated copy constructor for the different Tensor containers used in the code. Also there was no way to use off-the-shelf C2 operators in my code easily so I had to copy some code that is doing basic matmul, cat, split, transpose and linear as utility functions.
Two things missing:
- Profiling this implementation against the current ONNXified LSTM op
- Make this operator available to use in PyTorch
Reviewed By: dzhulgakov
Differential Revision: D14351575
fbshipit-source-id: 3b99b53212cf593c7a49e45580b5a07b90809e64
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/17461
Implementing a standalone LSTM Operator in Caffe2 adopted from this Aten implementation: diffusion/FBS/browse/master/fbcode/caffe2/aten/src/ATen/native/RNN.cpp. The most tricky thing in this exercise was that caffe2::Tensor has no copy constructor that made it necessary to implement a custom templated copy constructor for the different Tensor containers used in the code. Also there was no way to use off-the-shelf C2 operators in my code easily so I had to copy some code that is doing basic matmul, cat, split, transpose and linear as utility functions.
Two things missing:
- Profiling this implementation against the current ONNXified LSTM op
- Make this operator available to use in PyTorch
Reviewed By: dzhulgakov
Differential Revision: D14160172
fbshipit-source-id: c33e3f9e8aeae578b64d97593cb031a251216029
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/15366
swap the old implementation with a slightly easier one to understand
I ran the tests and compared the number of chains compared to the old algorithm. This one outperforms on every test, but we have yet to see if that impacts performance at all.
old chain 34 nomnigraph chain 25
old chain 46 nomnigraph chain 34
old chain 228 nomnigraph chain 188
old chain 397 nomnigraph chain 338
Reviewed By: ilia-cher
Differential Revision: D13057451
fbshipit-source-id: ccd050bfead6eb94ab9c7b0a70b09a22c2b9e499
Summary:
This will let us install tests and other Caffe2 python code as a part of running Caffe2 tests in PyTorch.
Broken out of https://github.com/pytorch/pytorch/pull/13733/
cc pjh5 yf225
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14898
Reviewed By: pjh5
Differential Revision: D13381123
Pulled By: orionr
fbshipit-source-id: 0ec96629b0570f6cc2abb1d1d6fce084e7464dbe
Summary:
Goal of this PR is to unify cuda and hip device types in caffe2 python front end.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14221
Differential Revision: D13148564
Pulled By: bddppq
fbshipit-source-id: ef9bd2c7d238200165f217097ac5727e686d887b
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/13825
async_polling was an intermediate step towards async_scheduling and is not used
Reviewed By: yinghai
Differential Revision: D13019059
fbshipit-source-id: eee6ba53e7f476ddb481afba3bf1768303864d32
* add opencl + fpga context
adds an opencl context inside caffe2/fb which can be used for fpga access
* [Caffe2] Force tensor inference checks to be triggered during testing
We've started to rely on TensorInference functions more for different analysis. This diff ensures that the TensorInference function's result matches what is expected from the definition of the operator.
* Enable building //caffe2:torch with @mode/opt
In @mode/opt, python runs out of a PAR, which breaks a lot of
assumptions in the code about where templates/ folders live relative
to __file__. Rather than introduce hacks with parutil, I simply turn
template_path into a parameter for all the relevant functions and
thread it through from the top level.
* [Caffe2] Fix cost models for DotProduct and Div. Update Tensor Inference for dot product
As title. DotProduct states that output is a 1-D tensor (https://caffe2.ai/docs/operators-catalogue.html#dotproduct) though code suggests it is either 0- or 1-D depending on inputs. TensorInference defined to support implementation.
* [SG-MoE] Add an option to make the experts NOT as components
* [nomnigraph] Rename and fixup convertToNeuralNetOperator API
This will make things a bit cleaner
* no longer symlink THNN.h and THCUNN.h
* forced decoder network (onnx export)
Closes https://github.com/pytorch/translate/pull/95
Add networks in ensemble_export.py to create a forced decoding network from PyTorch NMT checkpoints. This network takes an arbitrary numberized (source, target) pair and returns the model score for the translation, including penalties.
Vocabulary reduction networks are also supported, but note that target indices which are not in the possible_translation_tokens generated for the source input will be trea
* Revert schema change to fix production models
Revert schema change to fix production models
* MockLogDeviceReader - rebase on FIX
# Goal
1), Build a make_mock_log_device_reader using make_mock_reader
2), Replace the real log_device_reader here: https://fburl.com/raihwf1p
# Log by D8151734
Real log_device_reader:
```
I0529 20:29:05.373108 954994 tensor.h:839] Tensor print_net/log of type std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >. Dims: (): read_net/ParseOpenTrainingRow:0
I0529 20:29:05.373244 954994 tensor.h:839] Tensor read_net/ParseOpenTrainin
* [C2/D2][1/n]: Nonnegative-Constrained Optimization -- log barrier
implement log barrier as a regularization method
* Add teacher weight screening.
Add teacher weight sceening according to teacher labels. If teacher label is zero, we do not use the distill loss in the objective function.
* Add NormalizerContext
See task for more detail. This implementation is a copy of what exists for RegularizerContext except for how the parameters are defined in the model_definition thrift file.
I'll try an alternative implementation which overrides the default arguments of functions instead like for argscopes in tensorflow.
https://github.com/pytorch/pytorch/compare/master...MaximeBoucher:update-from-facebook-0939578c068c?expand=1
* Adding cosine similarity option in dot processor
Add pairwise cosine similarity option in dot product.
Add an option to concate dot product and cosine similarity.
Add test cases.
* [nomnigraph][redo] Concat elim for sparseNN
Same as D7962948, which was reverted because Operator Schema was not
defined
* [pytorch] Revert pytorch/pytorch#7918 'Release GIL when copying to shared memory', breaks ASAN
Revert this pytorch diff that breaks ASAN when running Filament in dev mode; in opt mode it gives "bad file descriptor" errors. Looks like a race when copying tensors to shared memory in multiple mp.Queue's (which spawn separate threads).
https://github.com/pytorch/pytorch/pull/7918/files
* [nomnigraph][mobile] Enable nomnigraph by default, use -Oz on nomnigraph related code to reduce code size
enables nomnigraph and reduces codesize
* [Warmup] Allow both offline incremental training and online training
Change plan name on saving side and reading side to support both training type
This diff depends on D8128530 and D8168651.
* Revert D7802642: [Warmup] Allow both offline incremental training and online training
This reverts commit afc213cf9b36cecf75333a788391c4d09f4afccc
@bypass-lint
An infra SEV is better than not reverting this diff.
If you copy this password, see you in SEV Review!
@cause_a_sev_many_files
* Add legacy grad logic to fix div op on old graphs.
Add legacy grad logic to fix div op on old graphs.
* Correctly propagate operator failures
Propagate errors from operators that throw exceptions and return false
* Revert D8374829: [caffe2][nomnigraph][redo] Concat elim for sparseNN
This reverts commit 6dda028c463e54bb5c32188bbbe9202107e188a5
@bypass-lint
An infra SEV is better than not reverting this diff.
If you copy this password, see you in SEV Review!
@cause_a_sev_many_files
* [Caffe2] Added extra_info to core.DeviceOption(), enforced extra_info to be inherited in scope.DeviceScope
extra_info is a newly defined field in DeviceOption proto. This diff added extra_info to the core.DeviceOption(). And, In scope.DeviceScope(), this diff enforce the new scope to inherit the extra_info from old scope.
* [opt] hgdirsync wasn't enabled, merge diverged code
Here's the damage, P59732616 basically xplat was left behind but had
the change from assert to CAFFE_ENFORCE
* OMP parallelism over RoIs for RoIAlign op
Simpler to parallelize over RoIs. Shouldn't affect other uses as it relies on
the number of OMP threads set during startup.
PR: https://github.com/pytorch/pytorch/pull/8562
* Use int64_t for shape in FillOps
to avoid overflow of int32
* Implement Rotated RoIAlign op
Based on Rotated RPNs as explained in https://arxiv.org/abs/1703.01086.
The idea is simple - orientation/angle is added as an RPN
anchor parameter and then the angle is further regressed similar to bbox
coords. There are some additional changes related to NMS and IoU, but besides
that it's a direct extension to Faster-RCNN. Further details in https://fb.quip.com/sZHlA1iMfWPZ.
RoIs are represented in [center_x, center_y, width, height, angle] format.
`angle` repre
* Rotated RoIAlign op CUDA forward implementation
CUDA forward impl for D8415490
* RoIAlignRotated op CUDA backward pass implementation
TSIA
* All remaining fixes to eliminate process_github.sh
Most of this diff has already been reviewed separately, except for the parts relating to _thnn/utils.py and _utils._internal.py
remove skipIf(True, 'Fbcode') line from process_github.sh
replace sed of cpp file with #ifdef to control cudnnDestroy use
undo sync-time deletion of .gitattributes, remove process_github.sh
switch to using _utils._internal rather than try-import-except
This diff also fixes the open-source bug where rebuilds have
* Back out "Revert D7802642: [Warmup] Allow both offline incremental training and online training"
Original commit changeset: 7707d2efe60e The original diff is backout becuase the online trainer package is backed out. This code would only work with new online trainer package
* [easy] improve error log in adagrad op
as title
* re-allow use of thnn_h_path
This fixes cffi usage in OSS
* [4/4] [tum] paralyzing layerNorm for GPU full sync
as title
* add compile=False to pytorch tests, remove hack with pyc
* Add shape and type inference for RowWiseArgMax operator
See title
* Revert D8515341: Back out "Revert D7802642: [Warmup] Allow both offline incremental training and online training"
This reverts commit 78167eeef0af16b60f72c82f9dcdda9b41b4dcbd
@bypass-lint
An infra SEV is better than not reverting this diff.
If you copy this password, see you in SEV Review!
@cause_a_sev_many_files
* [fix-flaky-test] mock_hive_reader_test flaky, because GlobalCounter collects local counts intervally
# Problem
`MockHiveReader` uses `GlobalCounter` to limit `max_examples`.
GlobalCounter on server node collect local counts from worker nodes every 1 sec.
This 1 sec delay makes it impossible to limit exactly to the `max_examples`, it will definitely exceed `max_examples`.
# Plan
Given,
```
Expected num_examples = max_examples + num_examples/sec (Read Speed) x 1 sec (GlobalCounter Sync Int
* [Caffe2] Fix FCGradient cost inference. Prevent overflow in cost inference
FCGradient missed a factor 2 in the `num_outputs == 3` case. Overflow was occurring with flop calculation for FC. Changed types to `uint64_t` to prevent future problems.
* Fix binary ops with empty inputs
Fix binary ops with empty inputs
* Support the filling of input blob with provided data
as title for Biz Integrity case
* Back out "Revert D8515341: Back out "Revert D7802642: [Warmup] Allow both offline incremental training and online training""
Original commit changeset: 30c55dd38816 Original diff is reverted due to introducing bad integration test. Fixed the integration test.
* [c2][easy] improve pack ops error loggings
as desc.
* Add ShapeTypeInference for LpNorm operator
As desc
* Shard test_nn to reduce runtime for each test target
Closes https://github.com/pytorch/pytorch/pull/8793
The current test_nn would time out and be disabled in GreenWarden, and we need to have an option to split it up in order to pass the stress test. Right now GreenWarden roughly allows running 100 test cases in test_nn before timing out, and here we have an option to divide test_nn into 30 shards (with ~40 tests in each shard) to allow for some test suite growth in the future.
* Change default caffe2_streams_per_gpu to 1
* Remove IN_SANDCASTLE from common.py and test_nn.py
We prefer to disable the failing tests through Sandcastle UI instead.
* Add a new class for an updated prof_dag.proto
This diff contains:
- An updated prof_dag.proto that contains blob profiles.
- A class to deserialize this information (serialization is in a follow up diff)
- Update to separate profiling information from NeuralNet (and use it as part of the class above).
- Unit tests
* Lambdarank for SparseNN
This diff adds a lambda_rank_layer for SparseNN.
changes include
1) Adds support for multi sessions in c2 op
2) Adds support for two different loss functions in c2 op
3) Unit tests for op
* Revert D8586950: Back out "Revert D8515341: Back out "Revert D7802642: [Warmup] Allow both offline incremental training and online training""
This reverts commit 012220ed63eccc35659a57b31d16a3625da6317b
@bypass-lint
An infra SEV is better than not reverting this diff.
If you copy this password, see you in SEV Review!
@cause_a_sev_many_files
* [easy] A few fixups to multithread predictor benchmark
(1) support perf on T6 server
(2) remove dead code
* fix a bug about the map size
as title
* Fix reduce sum on in-place case.
Fix reduce sum on in-place case.
* [Warmup] Reland reverted diff Allow both offline incremental training and online training
Closes https://github.com/pytorch/pytorch/pull/8827
fix net transform integration test. Allow offline and online trainer to coexist D7802642.
* Add StoreHandlerNotAvailableException
Add an exception for a store that is not available or has been
deleted.
* Use exception handling for fault tolerance, missing KV store
Remove status blobs to communication ops so that exceptions propagate on
failure.
* [C2/D2][2/n]: Nonnegative-Constrained Optimization -- bounded grad proj
for simple bounded constrained optimization, incl non-negative box constraints.
* [GanH]: Adaptive Weighting with More Estimations
With implemented postivity optimization, we now learn adaptive weights with different
parameterizations.
This improves parameter estimation and training stability.
* Revert some changes for landing
* Remove AutoNoGIL in StorageSharing
* Temporarily disable net_tests
* Revert "[Caffe2] Force tensor inference checks to be triggered during testing"
This reverts commit 67ef05c22b2f71b4a489695384932f968384a2a4.
* Revert "Fix reduce sum on in-place case."
This reverts commit 6cb8a8e1b3db7b6d20941b0053e3f3836068eb64.
* Revert "Revert "Fix reduce sum on in-place case.""
This reverts commit 130a257c0893dc09f4bd6e6a45d112261807fd2c.
Summary: Make test less computationally expensive
Reviewed By: Yangqing, dzhulgakov
Differential Revision: D6766236
fbshipit-source-id: 59e51faa1331d804b11da9f7237ee9ce0cb27df8
Summary:
Async executor based on async_polling (D5985110):
- Tasks scheduling other tasks, using polling only when necessary (e.g.
CUDA->CPU case)
- Fully async, i.e. RunAsync immediately returns
Reviewed By: azzolini
Differential Revision: D6281681
fbshipit-source-id: 06e3723e1424ffab652c38ca7b279cf76e43fa44
Summary:
Implementation of polling async net executor.
Notes:
- New net executor async_polling - schedules CPU and GPU ops asynchronously, uses single polling thread
- Events: update to Caffe2 events to support async CPU events, adding new methods:
Query() - non-blocking checking of event states: INITIALIZED -> RECORDED -> SUCCESS/FAILED
ErrorMessage() - when operation runs asynchronously and fails calling this on event will give error message
- Tasks: using existing DAGNet's algorithm to compute CPU and GPU chains, a separate task for each chain
- Polling: using single thread to query state of events - for CPU tasks atomically queries task state, for GPU task - uses cudaEventQuery; using Event
- Scheduling of CPU ops: using global thread pools
- Scheduling of GPU ops: using GPU thread pool per GPU device
Reviewed By: dzhulgakov
Differential Revision: D5985110
fbshipit-source-id: a9de7fcbb71d046a3aa1b573072b89a65dfeee8c
Summary: Adding ability to reuse workspace in Do op and unit tests
Reviewed By: akyrola
Differential Revision: D6037992
fbshipit-source-id: 73d6a14001f667f7ca5e1e02ff39911dc65e4cd1
Summary: We have to use copy constructor in Concat when copying non-primitive types
Reviewed By: Yangqing
Differential Revision: D6002883
fbshipit-source-id: 0aebc955079975bb6423291589ed09ce0660acf3
Summary: Use only MLP model and re-enable test
Reviewed By: bddppq, Yangqing
Differential Revision: D6013471
fbshipit-source-id: 0cb4a9346c62a739ee6259832181f71e60eef311
Summary:
Executor benchmarks to measure QPS for different models (sparse nn hogwild and
dataparallel, resnet50 dataparallel)
Reviewed By: dzhulgakov
Differential Revision: D5950770
fbshipit-source-id: 9aa8e0480468a55a6a97b10589d785c682fae01e
Summary: Adjust test thresholds and number of examples
Reviewed By: salexspb
Differential Revision: D5945588
fbshipit-source-id: 7aecb8c642d8775f51dd3c296a28f1faf7ae0c81
Summary:
Executor test that checks on different models that model params are the same
when using a given executor and simple net
Reviewed By: akyrola
Differential Revision: D5908769
fbshipit-source-id: b6f5a2cf89c5c67b68e8b9be3264f38d5740d897
