[DDP] Add PackedSequence support when device_ids is specified
#86614
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
[ghstack-poisoned]
awgu
requested review from
mrshenli,
pritamdamania87,
zhaojuanmao,
rohan-varma,
H-Huang,
kwen2501 and
mingzhe09088
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/86614
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 4d7c335: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
pytorch-bot
bot
added
the
release notes: distributed (c10d)
awgu
added a commit
that referenced
this pull request
Oct 10, 2022
ghstack-source-id: 6cdbf552770ec74c5295bd96fe14a80ec4cf3c30 Pull Request resolved: #86614
awgu
commented
Oct 10, 2022
| @@ -1,10 +1,12 @@ | |||
| from typing import Any, Dict, List, Tuple | |||
…ecified" Before this PR, if a user runs DDP with `device_ids` specified and with a `PackedSequence` input, then the execution will error with something like: ``` raise ValueError( ValueError: batch_sizes should always be on CPU. Instances of PackedSequence should never be created manually. They should be instantiated by functions like pack_sequence and pack_padded_sequences in nn.utils.rnn. https://pytorch.org/docs/stable/nn.html... ``` This is because the DDP forward calls `_to_kwargs()`, which calls `_recursive_to()`, which moves the inputs to GPU. However, `_is_namedtuple(packed_sequence)` returns `True`, leading to the branch `return [type(obj)(*args) for args in zip(*map(to_map, obj))]`, which tries to construct a `PackedSequence` directly via `type(obj)(*args)`, leading to the error. Repro for `_is_namedtuple(packed_sequence)` returning `True`: ``` import random import torch import torch.nn.utils.rnn as rnn_utils from torch.nn.parallel.scatter_gather import _is_namedtuple def _ordered_sequence(tensor_type): seqs = [tensor_type(random.randint(1, 256)) for _ in range(32)] seqs = [s.random_(-128, 128) for s in seqs] ordered = sorted(seqs, key=len, reverse=True) return ordered def _padded_sequence(tensor_type): ordered = _ordered_sequence(tensor_type) lengths = [len(i) for i in ordered] padded_tensor = rnn_utils.pad_sequence(ordered) return padded_tensor, lengths padded, lengths = _padded_sequence(torch.Tensor) packed = rnn_utils.pack_padded_sequence( padded, lengths, enforce_sorted=False) print(type(packed), packed.data.device) print(_is_namedtuple(packed)) ``` Test Plan: ``` python test/distributed/test_c10d_nccl.py -k test_ddp_packed_sequence ``` [ghstack-poisoned]
awgu
added a commit
that referenced
this pull request
Oct 10, 2022
ghstack-source-id: 4edd04038037ceea107d4946253583c369f8dde1 Pull Request resolved: #86614
awgu
added
module: ddp
…ecified" Before this PR, if a user runs DDP with `device_ids` specified and with a `PackedSequence` input, then the execution will error with something like: ``` raise ValueError( ValueError: batch_sizes should always be on CPU. Instances of PackedSequence should never be created manually. They should be instantiated by functions like pack_sequence and pack_padded_sequences in nn.utils.rnn. https://pytorch.org/docs/stable/nn.html... ``` This is because the DDP forward calls `_to_kwargs()`, which calls `_recursive_to()`, which moves the inputs to GPU. However, `_is_namedtuple(packed_sequence)` returns `True`, leading to the branch `return [type(obj)(*args) for args in zip(*map(to_map, obj))]`, which tries to construct a `PackedSequence` directly via `type(obj)(*args)`, leading to the error. Repro for `_is_namedtuple(packed_sequence)` returning `True`: ``` import random import torch import torch.nn.utils.rnn as rnn_utils from torch.nn.parallel.scatter_gather import _is_namedtuple def _ordered_sequence(tensor_type): seqs = [tensor_type(random.randint(1, 256)) for _ in range(32)] seqs = [s.random_(-128, 128) for s in seqs] ordered = sorted(seqs, key=len, reverse=True) return ordered def _padded_sequence(tensor_type): ordered = _ordered_sequence(tensor_type) lengths = [len(i) for i in ordered] padded_tensor = rnn_utils.pad_sequence(ordered) return padded_tensor, lengths padded, lengths = _padded_sequence(torch.Tensor) packed = rnn_utils.pack_padded_sequence( padded, lengths, enforce_sorted=False) print(type(packed), packed.data.device) print(_is_namedtuple(packed)) ``` Test Plan: ``` python test/distributed/test_c10d_nccl.py -k test_ddp_packed_sequence ``` Without the fix, the added unit test fails with the expected error. [ghstack-poisoned]
awgu
added a commit
that referenced
this pull request
Oct 10, 2022
ghstack-source-id: d7445464b164f702035912100a0caffe51c02df3 Pull Request resolved: #86614
…ecified" Before this PR, if a user runs DDP with `device_ids` specified and with a `PackedSequence` input, then the execution will error with something like: ``` raise ValueError( ValueError: batch_sizes should always be on CPU. Instances of PackedSequence should never be created manually. They should be instantiated by functions like pack_sequence and pack_padded_sequences in nn.utils.rnn. https://pytorch.org/docs/stable/nn.html... ``` This is because the DDP forward calls `_to_kwargs()`, which calls `_recursive_to()`, which moves the inputs to GPU. However, `_is_namedtuple(packed_sequence)` returns `True`, leading to the branch `return [type(obj)(*args) for args in zip(*map(to_map, obj))]`, which tries to construct a `PackedSequence` directly via `type(obj)(*args)`, leading to the error. Repro for `_is_namedtuple(packed_sequence)` returning `True`: ``` import random import torch import torch.nn.utils.rnn as rnn_utils from torch.nn.parallel.scatter_gather import _is_namedtuple def _ordered_sequence(tensor_type): seqs = [tensor_type(random.randint(1, 256)) for _ in range(32)] seqs = [s.random_(-128, 128) for s in seqs] ordered = sorted(seqs, key=len, reverse=True) return ordered def _padded_sequence(tensor_type): ordered = _ordered_sequence(tensor_type) lengths = [len(i) for i in ordered] padded_tensor = rnn_utils.pad_sequence(ordered) return padded_tensor, lengths padded, lengths = _padded_sequence(torch.Tensor) packed = rnn_utils.pack_padded_sequence( padded, lengths, enforce_sorted=False) print(type(packed), packed.data.device) print(_is_namedtuple(packed)) ``` Test Plan: ``` python test/distributed/test_c10d_nccl.py -k test_ddp_packed_sequence ``` Without the fix, the added unit test fails with the expected error. [ghstack-poisoned]
awgu
added a commit
that referenced
this pull request
Oct 10, 2022
ghstack-source-id: 78bab0549d3b65a8e2b4933ad07544d96f49f616 Pull Request resolved: #86614
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
|
Hey @awgu. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Stack from ghstack:
PackedSequencesupport whendevice_idsis specified #86614 [DDP] AddPackedSequencesupport whendevice_idsis specifiedutils.py(moved to_utils.py) #86528 [FSDP] Removeutils.py(moved to_utils.py)summon_full_params(with_grads=True)#85738 [FSDP] Add initialsummon_full_params(with_grads=True)use_orig_params#84911 [FSDP] Adduse_orig_paramsBefore this PR, if a user runs DDP with
device_idsspecified and with aPackedSequenceinput, then the execution will error with something like:This is because the DDP forward calls
_to_kwargs(), which calls_recursive_to(), which moves the inputs to GPU. However,_is_namedtuple(packed_sequence)returnsTrue, leading to the branchreturn [type(obj)(*args) for args in zip(*map(to_map, obj))], which tries to construct aPackedSequencedirectly viatype(obj)(*args), leading to the error.Repro for
_is_namedtuple(packed_sequence)returningTrue:Test Plan:
Without the fix, the added unit test fails with the expected error.