Commits

Branch selector

 main

User selector

mikaylagawarecki

Datepicker

All time

Commit History

Commits on May 31, 2024

• Update docstring for weights_only (pytorch#127575)

  Show description for 4644def
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  4644def

  Copy full SHA for 4644def

Commits on May 30, 2024

• Default meta device to use swap_tensors in nn.Module._apply (.to_empty and .to('meta')) (pytorch#126819)

  Show description for fa426b0
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  fa426b0

  Copy full SHA for fa426b0

• Default XLA to use swap_tensors path in nn.Module._apply (pytorch#126814)

  Show description for bfdec93
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  bfdec93

  Copy full SHA for bfdec93

• Relax use_count constraints for swap_tensors when AccumulateGrad holds a reference (pytorch#127313)

  Show description for cd06ae0
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  cd06ae0

  Copy full SHA for cd06ae0

Commits on May 24, 2024

• Fix map_location for wrapper subclass and device tensors that go through numpy (pytorch#126728)

  Show description for 87f79af
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  87f79af

  Copy full SHA for 87f79af

Commits on May 23, 2024

• Default meta device to use swap_tensors in nn.Module._apply (.to_empty and .to('meta')) (pytorch#126819)

  Show description for ab61309
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  ab61309

  Copy full SHA for ab61309

• Default XLA to use swap_tensors path in nn.Module._apply (pytorch#126814)

  Show description for eb41ed5
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  eb41ed5

  Copy full SHA for eb41ed5

Commits on May 17, 2024

• Allow tensor subclasses and add torch.serialization.add_safe_globals that allows users to allowlist classes for weights_only load (pytorch#124331)

  Show description for 66dc8fb
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  66dc8fb

  Copy full SHA for 66dc8fb

• [easy] Fix typing for map_location docs in torch.load (pytorch#125473)

  Show description for 776b878
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  776b878

  Copy full SHA for 776b878

Commits on May 14, 2024

• Reland add write_record_metadata to PyTorchFileWriter (pytorch#126087)

  Show description for bbdbfe3
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  bbdbfe3

  Copy full SHA for bbdbfe3

Commits on May 3, 2024

• Add write_record_metadata to PyTorchFileWriter (pytorch#125184)

  Show description for dd92637
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  dd92637

  Copy full SHA for dd92637

Commits on Apr 30, 2024

• Add MAP_SHARED option for torch.load(mmap=True) (pytorch#124889)

  Show description for 2480e8b
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  2480e8b

  Copy full SHA for 2480e8b

Commits on Apr 23, 2024

• Add testing and fix weights_only load for quantized types and nn.Parameters with python attrs (pytorch#124330)

  Show description for c82fcb7
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  c82fcb7

  Copy full SHA for c82fcb7

Commits on Apr 18, 2024

• Add swap_tensors path to nn parametrizations (pytorch#124130)

  Show description for 5ba6bb7
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  5ba6bb7

  Copy full SHA for 5ba6bb7

Commits on Apr 17, 2024

• Add swap_tensors path to nn parametrizations (pytorch#124130)

  Show description for 64f6ddf
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  64f6ddf

  Copy full SHA for 64f6ddf

Commits on Apr 16, 2024

• Add testing and fix issues for weights_only load for LRScheduler (pytorch#123775)

  Show description for 383d2d1
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  383d2d1

  Copy full SHA for 383d2d1

Commits on Mar 29, 2024

• Add RMSNorm module (pytorch#121364)

  Show description for 487b6d4
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  487b6d4

  Copy full SHA for 487b6d4

Commits on Mar 27, 2024

• Fix swap_tensors path in _apply for modules that inherit from RNNBase (RNN, GRU, LSTM) (pytorch#122800)

  Show description for cc12668
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  cc12668

  Copy full SHA for cc12668

• Add RMSNorm module (pytorch#121364)

  Show description for a7306de
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  a7306de

  Copy full SHA for a7306de

Commits on Mar 6, 2024

• Add assign argument to torch.Tensor.module_load (pytorch#121158)

  Show description for 4b39033
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  4b39033

  Copy full SHA for 4b39033

• [easy] Fixed requires_grad preservation for nn.Module.load_state_dict(assign=True) (pytorch#121157)

  Show description for 27389e0
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  27389e0

  Copy full SHA for 27389e0

Commits on Mar 5, 2024

• Add exhaustive module and optimizer tests for torch.load(state_dict, weights_only=True) (pytorch#121049)

  Show description for d621e3e
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  d621e3e

  Copy full SHA for d621e3e

Commits on Feb 28, 2024

• Update nn.Module._apply to not gate on should_use_set_data when swap_tensors is set (pytorch#120659)

  Show description for 677e67c
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  677e67c

  Copy full SHA for 677e67c

Commits on Feb 9, 2024

• Integrate swap_tensors into nn.Module.load_state_dict (pytorch#117913)

  Show description for 3372aa5
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  3372aa5

  Copy full SHA for 3372aa5

Commits on Feb 8, 2024

• [BE] Add dtypesIfMPS to ModuleInfo enabling float16 tests for MPS and remove all skipIfMPS for float64 (pytorch#119039)

  Show description for db1a4dc
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  db1a4dc

  Copy full SHA for db1a4dc

Commits on Feb 7, 2024

• [easy] Add testing utilties for torch.nn.utils.set_swap_module_params_on_conversion (pytorch#118023)

  Show description for 23b030a
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  23b030a

  Copy full SHA for 23b030a

• Add swap_tensors path to nn.Module._apply (pytorch#117167)

  Show description for d5a718d
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  d5a718d

  Copy full SHA for d5a718d

Commits on Feb 5, 2024

• [BE] Add dtypesIfMPS to ModuleInfo enabling float16 tests for MPS and remove all skipIfMPS for float64 (pytorch#119039)

  Show description for 04d52d5
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  04d52d5

  Copy full SHA for 04d52d5

• Move nn.Module.load_state_dict tests from test_nn.py to separate file (pytorch#118028)

  Show description for b92819a
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  b92819a

  Copy full SHA for b92819a

Commits on Feb 3, 2024

• Document file format returned by torch.save (pytorch#118719)

  Show description for 9ffed22
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  9ffed22

  Copy full SHA for 9ffed22

Commits on Feb 1, 2024

• Add myself to CODEOWNERS for serialization-related files (pytorch#118892)

  Show description for 39aab55
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  39aab55

  Copy full SHA for 39aab55

• [easy] Enable test_neg_view for 5D SampleInput for torch.nn.functional.linear (pytorch#118815)

  Show description for 45d2dff
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  45d2dff

  Copy full SHA for 45d2dff

Commits on Jan 29, 2024

• [easy] Fix small typo in register_state_dict_pre_hook doc (pytorch#118571)

  Show description for 796d270
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  796d270

  Copy full SHA for 796d270

Commits on Jan 24, 2024

• Fix swap_tensors to swap PyObjects associated with TensorImpl (pytorch#116955)

  Show description for 41a56f7
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  41a56f7

  Copy full SHA for 41a56f7

Commits on Jan 5, 2024

• Fix TransformerEncoderLayer for bias=False (pytorch#116760)

  Show description for d0cf218
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  d0cf218

  Copy full SHA for d0cf218

Pagination

Previous

Next