Comments (2)
ezyang
commented on May 10, 2024
2
I strongly prefer (1) too, we wouldn't even be having this discussion if we always inlined into NN modules.
also cc @anijain2305 on nn module inlining
from pytorch.
albanD
commented on May 10, 2024
1
Agreed that (1) sounds like the way to go.
Also I would expect that this is what would happen "naturally" if Dynamo was inlining the @property and nested module calls inside the python implementation of parametrization.
from pytorch.
Related Issues (20)
- DISABLED test_quantization_doc_fx (__main__.TestQuantizationDocs) HOT 2
- DISABLED test_view_and_inplace_view (__main__.TestAOTAutograd) HOT 1
- [inductor][cpu]mobilenet_v2_quantized_qat float32 single thread static/dynamic shape CPP/default wrapper performance regression in 2024-04-28 nightly release
- [BUG]Nan in gradients of scaled_dot_product_attention operation with mem_efficient backend
- Unnecessary warning when numpy not installed
- [RFC] Add Cpp Template for GEMM related ops via max-autotune for Inductor CPU
- MAX-Autotune Compilation Time Regression Due To Added MM Configs HOT 1
- cnm
- DISABLED [WORKFLOW_NAME] / [PLATFORM_NAME] / [JOB_NAME] HOT 1
- cnm
- [Dynamo] Support tracing through _get_current_dispatch_mode_stack
- Have config/env option to disable all PT2 caching
- [dynamo] fix nn.Module @property that accesses closure cells
- KINETO_USE_DAEMON causing issues
- `torch.compile` and complex numbers HOT 2
- Support dynamo tracing weakref obj
- Migrate multiple/custom runner labels before deprecation
- torch._inductor.config.max_autotune_gemm_backends = "TRITON" crashes with Convolution layer
- ☂️ `torch.compile` generates slower code for LLMs than eager on ARM platform (M1/AARCH64)
- [ARM] `Vectorized<half>::loadu(x, 8)` yields slow code if `-fno-unsafe-math-optimizations` are used 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.