Comments (1)
When training BERT with compiled autograd on multiple GPUs, the propose_partition
method took about 3 minutes to execute. After applying this proposed change (linked PR), the execution time was reduced 10x, i.e., it took about 17 seconds.
from pytorch.
Related Issues (20)
- CLA required despite being existing contributor HOT 2
- DISABLED test__int_mm_k_32_n_16_use_transpose_a_True_use_transpose_b_False_cuda (__main__.TestLinalgCUDA) HOT 1
- Python sometimes crashes inexplicably, with "/var/log/kernel. log" displaying index errors HOT 3
- torch.linalg.inv returns constantly the identity matrix for input on GPU HOT 1
- torch.fx.symbolic_trace doesn't support many Callable types
- [BUG][JIT] `torch.jit.script` is not compatible with `DeprecationWarning` and `FutureWarning` HOT 9
- Fused AdamW maybe should accept lr_dict directly? HOT 1
- Overly demanding PyTorch CLA HOT 1
- UNSTABLE linux-binary-manywheel / manywheel-py3_8-cuda12_1-test / test HOT 2
- UNSTABLE linux-binary-manywheel / manywheel-py3_8-cuda12_4-test / test HOT 2
- I donβt know if itβs a problem with cuda or pytorch HOT 3
- torch.compile (inductor) bug random signed number generation HOT 6
- [While_loop] How to use layer like `torch.nn.BatchNorm2d` with while_loop? HOT 8
- torch.compile reorder_for_compute_comm_overlap sink_waits pass does not work HOT 2
- Inductor fails at assert self.symbol_to_source.get(expr) HOT 4
- Add option for custom ops to automatically get a FakeTensor kernel (during static shapes) HOT 1
- DISABLED test_arange2_dynamic_shapes_cuda (__main__.DynamicShapesGPUTests) HOT 1
- DISABLED test_comprehensive_fft_ifft_cuda_float64 (__main__.TestInductorOpInfoCUDA) HOT 1
- ~PyTorch Docathon H1 2024!~
- Dynamo should prune non-live captured variables
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
π 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. πππ
-
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 β€οΈ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from pytorch.