- 29 Jul, 2022 2 commits
-
-
hubertlu-tw authored
-
hubertlu-tw authored
-
- 26 Jul, 2022 1 commit
-
-
Tim Moon authored
* Hack to enable training GPT-3 Seems to fix bug from #1416 * Add test to initialize model-parallelism for decoder-only Transformers Namely GPT-3.
-
- 25 Jul, 2022 1 commit
-
-
Aidyn-A authored
-
- 20 Jul, 2022 1 commit
-
-
Aidyn-A authored
* add test * update batch sizes * update batch sizes * small updates * delete comment * add async comm * add sync if needed * update tests * remove redundant imports * code cleanup * minor updates * update dtype for comparison * fix dtypes * fix typo * modify sizes and use common_utils.find_free_port * fix typo and use double precision * revert some changes, create test for profiling on L1 * remove redundant line * revert UCC_TLS and add sync to fwd_bwd * code clean up * code clean up * modify BERT test * add comment
-
- 14 Jul, 2022 1 commit
-
-
Sandeep Subramanian authored
* Time dimension shape check for fused scale mask softmax kernel Signed-off-by:
MaximumEntropy <sandeep.subramanian.1@umontreal.ca> * Add shape test Signed-off-by:
-
- 11 Jul, 2022 1 commit
-
-
Perkz Zheng authored
* update: mpu for t5 rpe * update: add rpe mpu group test * fix semicolon bugs Co-authored-by:
Masaki Kozuki <masaki.kozuki.2014@gmail.com> * fix semicolon bugs Co-authored-by:
-
- 07 Jul, 2022 1 commit
-
-
Masaki Kozuki authored
* remove pyprof * remove reparameterization * remove pyprof test * clean up
-
- 23 Jun, 2022 2 commits
-
-
Masaki Kozuki authored
* it looks possible to remove this file * add communication collectives * update Column|RowParallelLinear * update checkpoint function * update function name * parity between public and private collectives * row parallel linear * column parallel linear * sequence parallel: p2p comm fix typo * sequence parallel: pipeline parallel * fix typo * add layernorm with sequence_parallel_enabled attr * class variable -> member variable * fix col parallel test with sequence parallel * Initial test of `forward_backward_pipelining_without_interleaving` with `model_type=ModelType.encoder_and_decoder` * add cases pretending to test sequence_parallel * Apply 2 suggestion(s) to 1 file(s) * update sequence_parallel_enabled docstring * update docstring: order of tensor dimensions, sequence_parallel_enabled behavior * Divide sequence_length if sequence parallel tensor shape should be updated if sequence parallel is enabled. * cherry-pick https://github.com/NVIDIA/Megatron-LM/commit/8474e6e54fcb9dfa37aea039352f9fb485fb6f61 * type annotation * Fix matmul call in RowParallelLinear Fix `sequence_parallel_enabled` to `False` as you can see in https://github.com/NVIDIA/Megatron-LM/blob/d898a8991d1a08d29074f87819d1bf41517e35f5/megatron/mpu/layers.py#L511-L514 * update rowparallellinear test * fix `loss_weight` is not defined in test_layers * @eqy's comment * mixed fused layer norm * fix typo * misc * test_layers cleanup * Skip Bert/GPT script Since these two models haven't gotten updated for sequence parallle, e.g. the update of the order of dimension from (batch, sequence, feature) to (sequence, batch, feature) and global variables of arguments * debug part 1/N: comment out `x.retain_grad` * debug part 2/N: [ColumnParallelLinear] comment out overriding of sequence_parallel_enabled * debug 3/N: add pipeline test with parallel mlp * Fix handling `self.input_tensor` and argument * tp2pp4 ModelType.encoder_or_decoder is failing, which can be at my fault because the backward is blaming the output and the grad_ouptut shape don't match * revert debug 1/N * defer tensor model parallel size > 1 * split tensor in sequence dim * cosmetic * cosmetic: remove archaic comment * enable TP>1 for encoder_and_decoder as well * set requires_grad=True always... * Set `scatter_gather_tensors_in_pipeline` to :obj:`False` for the sake of nemo megatron's GPT works with sequence parallel enabled. * brush up comment of `requires_grad()` There's a possibility that PyTorch DistributedDataParallel hangs when some tensor (or parameter) doesn't require grad according to @ptrblck. This forced `requires_grad` in my understanding is different from that. * misc changes of scatter_gather_tensors_in_pipeline comment * guard for torch_ucc * cosmetic changes related to tests * update command line arguments * update TransformerLanguageModel * rename * move gpt to gpt.py * update bert * add all_gather for params in sequence parallel region * misc. some diffs were lost during rebasing... * updates for non sequence parallel execution * gpt with sequence parallel * Apply 2 suggestion(s) to 2 file(s) * update tensor&pipeline parallel size * why `sequence_parallel_enabled` is not supplied!? Did I messed up when rebasing? * cosmetic fix * correct key is sequence_parallel_enabled
-
Tim Moon authored
* Increase default bucket size in distributed Adam * Move distributed Adam unit test to contrib tests Integrate into unit testing framework * Tweak hyperparameters for dist Adam optimizer test Improves numerical stability so we can keep tight tolerances. Adopting suggestions from @crcrpar. * Use distributed test infrastructure in distributed Adam unit test Suggestion from @crcrpar.
-
- 22 Jun, 2022 1 commit
-
-
Masaki Kozuki authored
* add temporary dispatch of double, float, half, bfloat16 * fusedadam of bfloat16 * Add bfloat16 path to FusedAdam
-
- 14 Jun, 2022 2 commits
- 31 May, 2022 1 commit
-
-
eqy authored
Do pipeline parallelism tests in double because TF32 environment variables can be painful to manage across test suites (#1391) * check in * skip interleaved with 2 GPU * change type annotation * address comments thanks @crcrpar @Aidyn-A
-
- 20 May, 2022 1 commit
-
-
Aidyn-A authored
* add grad check * change assert * minor changes * revert unnecessary changes * suggested changes * fix tensor comparison * small changes
-
- 19 May, 2022 2 commits
-
-
eqy authored
* check in * type * cleanup * cleanup * fix function call * Apply suggestions from code review Co-authored-by:Masaki Kozuki <mkozuki@nvidia.com>
-
eqy authored
* check in * fancy context style Co-authored-by:
-
- 18 May, 2022 1 commit
-
-
Masaki Kozuki authored
* NcclDistributedTestBase * fix stupid mistake * add UCC test * add UCC backend * torch ucc tests * allows for UCC backend * Set `UCX_TLS` to `tcp,cuda_copy` & Use DDP iff it makes sense * Apply 4 suggestion(s) to 1 file(s) * mix&match NCCL & UCC * use both ucc&nccl in gpt * UCC for Pipeline Parallel, NCCL for the others * conditionally use ucc * make ucc guards more friendly * test raises when torch_ucc isn't available * Change to member variable from class variable Co-authored-by:
Aidyn Aitzhan <31858918+Aidyn-A@users.noreply.github.com> * pass async_comm to train, I mistakenly dropped it during the rebase * fix typo: functionality * Enable tensor parallel only when device count > 4 I want pipeline model parallel world size to be >= 4 because previously I saw GPT/BERT failing when only UCC is used. So I'm speculating that there's some gotcha around pipeline size of 4. * Add nvidia driver version guard Co-authored-by:
-
- 12 May, 2022 1 commit
-
-
eqy authored
* initial check in * fix * fix test * address some review comments and cleanup * fix * bookmark * fix sync placement to come before gather * similar fix for non-gather case * add async bert * update gpt minimal test * allow selection of default pp test * fix bert test * cleanup * cleanup
-
- 11 May, 2022 1 commit
-
-
Aidyn-A authored
* add loss comparison to test_pipeline_parallel_fwd_bwd * applied some suggested changes * update test_pipeline_parallel_fwd_bwd.py * update test_pipeline_parallel_fwd_bwd.py 2 * minor update * update test_pipeline_parallel_fwd_bwd.py 3
-
- 29 Apr, 2022 1 commit
-
-
eqy authored
* fix typo * Update test_pipeline_parallel_fwd_bwd.py
-
- 15 Apr, 2022 2 commits
-
-
hubertlu-tw authored
-
eqy authored
* FusedRMSNorm based on FusedLayerNorm * refactor duplicated kernels * delete comments * delete comments * cleanup * cleanup * cleanup, fixed clobbering forward_affine_mixed_dtypes * fix pybind naming and add MixedFused test * undo skipping * check elementwise_affine * Update tests/L0/run_fused_layer_norm/test_fused_layer_norm.py Oof, nice catch, thanks Co-authored-by:
-
- 07 Apr, 2022 1 commit
-
-
Masaki Kozuki authored
* add test * destroy model parallel was missing
-
- 25 Mar, 2022 1 commit
-
-
Masaki Kozuki authored
* try PyTorch custom TestCase class * revert * initial working example * update * data utils * fix imports * hardcode backend to nccl * fix signature * fix typo * mapping * set device * init * refactor x entropy * remove unused import & destroy model parallel * refactor random * fix test * remove migrated tests * refactor * init * separate affine weight init * init model parallel * split more * weight init fix part 1 * use cpu init for consistency btwn native and tensor parallel * black * add col parallel * use a 3D tensor of square matrix for column parallel linear * skip the failing cases * migrate layers test * pipeline parallel forward/backward * fix typo * fix typo * fix * fix pipeline world size * black * rm `run_pipeline_parallel_test` in favor of test_pipeline_parallel_fwd_bwd.py * stop logging * set log level * black * license and format * fix * skip tf32 as matrices are small * remove potentially inappropriate license * Apply suggestions from code review * remove `TODO` comment * `torch.testing.assert_allclose` -> `torch.testing.assert_close` * remove comment-outs * remote unused import * minor fix
-
- 26 Feb, 2022 1 commit
-
-
Masaki Kozuki authored
* fuse grad accumulation w/ weight grad Co-authored-by:
Sangkug Lym <slym@nvidia.com> * fp32 training path * not using *args, **kwargs * backward: moved the tensor dimension cnversion Co-authored-by:
-
- 25 Feb, 2022 1 commit
-
-
Masaki Kozuki authored
-
- 23 Feb, 2022 1 commit
-
-
Masaki Kozuki authored
-
- 04 Feb, 2022 1 commit
-
-
eqy authored
* FusedRMSNorm based on FusedLayerNorm * refactor duplicated kernels * delete comments * delete comments * cleanup * cleanup * cleanup, fixed clobbering forward_affine_mixed_dtypes * fix pybind naming and add MixedFused test * undo skipping * check elementwise_affine * Update tests/L0/run_fused_layer_norm/test_fused_layer_norm.py Oof, nice catch, thanks Co-authored-by:
-
- 31 Jan, 2022 1 commit
-
-
Masaki Kozuki authored
* Free output tensor on each pipeline stage for smaller memory footprint see: https://github.com/NVIDIA/Megatron-LM/commit/057b086c689b164864455430c223ab52fd86bbcb * ref: https://github.com/NVIDIA/Megatron-LM/commit/945ece943149b63511e9d0ec3df8effe7f3c13ff * ref: https://github.com/NVIDIA/Megatron-LM/commit/9a8b89acd8f6ba096860170d0e30ddc0bc2bacd4 * remove position embedding group in destroy * pass deallocate_pipeline_outputs to backward_step * fix typo * missing deallocate_pipeline_outputs * fix typo: grad_ouptut -> grad_output * update tests * remove accessed todo * test with data parallel size of 2 if there's equal to or more than 8 gpus
-
- 28 Jan, 2022 2 commits
-
-
Masaki Kozuki authored
* cosmetic refactor in test * log with PID * log more info: rank, pid, filename, lineNo
-
Masaki Kozuki authored
* have get_kth_microbatch deal with None batch * broadcast based on tensor parallel rank * dtype * remove unnecessary .cuda() Processes of tensor parallel rank != 0 doesn't need to prepare one or more `torch.utils.data.DataLoader` instances, which means the argument of `batch` of `get_kth_microbatch` function can be `None` but the current function implementation doesn't allow for it.
-
- 21 Jan, 2022 1 commit
-
-
Masaki Kozuki authored
* add keyword argument of `grad_scaler` * update test * pass dtype to fwd_step_func * add log * calc loss in autocast as per https://pytorch.org/docs/stable/amp.html#autocasting * add keyword argument of `grad_scaler` * update test * pass dtype to fwd_step_func * add log * calc loss in autocast as per https://pytorch.org/docs/stable/amp.html#autocasting * option to turn off autocast inside forward_step function As there's some users who activate `autocast` outside fwd/bwd functions. * add missing arg of disable_autocast * reorder args of no pipeline
-
- 17 Dec, 2021 1 commit
-
-
Masaki Kozuki authored
Add an argument of `dtype` to forward_backward functions to specify the dtype used in p2p comm (#1249) * let users sepcify dtype for p2p comm taking the possibility of O2 style AMP into account * add `dtype` argument to forward_backward functions * fix * better message * add docstring of dtype * add a link to dtype logic of p2p comm
-
- 16 Dec, 2021 1 commit
-
-
eqy authored
* reduce bert memory usage, placeholder data for gpt * update gpt test * fix * Update tests/L0/run_transformer/run_bert_minimal_test.py remove debugging indexing Co-authored-by:
-
- 14 Dec, 2021 3 commits
- 10 Dec, 2021 2 commits
-
-
Masaki Kozuki authored
* update parallel_state * update pipeline common funcs - forward_step and backward_step * update pipelining w/o interleaving * type hint * merge utils into without_interleaving Motivation: functions in utils are only used by forward_backward_pipelining_without_interleaving * fix handling of `model_type` * fix import of DDP * update set_input_tensor method * fix * cosmetic * update model * refactor pipeline test scripts
-
Rishi Puri authored
Minimal gpt pipeline parallel (builds off of minimal_bert_pipeline_parallel) including cpu-offloading (#1222) * minimal bert pipeline parallel test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * first draft of gpt minimal test * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * framework to scale up the gpt2 test for variety of distributed setups * adding gpt_minimal_test to list of multigpu tests Co-authored-by:
Eddie Yan <eddiey@nvidia.com> Co-authored-by:
riship <riship@nvidia.com>
-
