Comments (4)
@pytorchbot label "oncall: distributed"
from pytorch.
@mvpatel2000 You can try #125339 to see if that PR can help. But it only supports DTensor.
from pytorch.
@mvpatel2000 Curious, how many nodes/ranks are you using? Also, is the state_dict/optimizer state_dict DTensor based? I'm also implement the broadcasting feature at this moment. But it will only support DTensor based state_dict/optimizer state_dict.
from pytorch.
@mvpatel2000 Curious, how many nodes/ranks are you using? Also, is the state_dict/optimizer state_dict DTensor based? I'm also implement the broadcasting feature at this moment. But it will only support DTensor based state_dict/optimizer state_dict.
This is for multiple GPUs (2 is sufficient).
We currently unit test this with ShardedTensors (as this is the default unless a device mesh is passed in), but I believe it's fine if it's only DTensor
from pytorch.
Related Issues (20)
- [export] Failed to trace HF Llama2 model HOT 3
- _foreach_atan2_ HOT 1
- torch.prod cannot be used with cudagraphs HOT 1
- Compilation time for dynamic shapes for gpt-fast prefill regressed significantly. HOT 2
- Correctness bugs and precision loss with Python-native-floating-point-number/single-valued-tensor divisions HOT 1
- [RFC] A device-agnostic Python runtime API design for stream-based accelerators HOT 2
- A Permute Layer for torch.nn.Sequential
- Forward hooks not called when fast path is used in TransformerEncoderLayer HOT 1
- How to enable XNNPACK instead of NNPACK/MKLDNN in Windows?
- torch._dynamo.exc.Unsupported: call_method GetAttrVariable(UnspecializedNNModuleVariable(CenterCrop), _transformed_types) __iter__ () {}
- AOTriton Cmake error breaking PyTorch nightly binary builds for ROCm
- `torch.compile` with `reduce-overhead`: very long compile time + GPU memory continuously to grow HOT 6
- warnings.warn is super spammy under Dynamo HOT 5
- taking upper triangular of "-inf" matrix results in nan values HOT 1
- Using a warning inside of Dynamo internals is super spammy
- Using PyTorch with Transformers to run inference with 'MPS' backend causes poor results.
- [v.2.4.0] Release Tracker HOT 15
- 'torch.compiler.reset()' does not reset 'assume_constant_result' value HOT 1
- [Dynamo] FunctionCtx initilaization warning suppression looks fishy
- HSDP + `set_optimizer_state_dict` errors with monolithic checkpointing HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pytorch.