Comments (3)
Agreed.
I'd actually go a step further and say you could dramatically simplify your architecture, while improving runtime speed, by using XLA as the only backend.
(You probably know that XLA was invented for exactly this purpose, to unite the backends of TensorFlow.)
I've used XLA (via JAX) quite a bit, and in my experience the code it generates is at least as fast as the code from other frameworks.
Of course this makes sense, because XLA typically has static access to the full graph including tensor shapes, so it has much more information at its disposal than e.g. PyTorch in eager mode.
For example, it can statically allocate tensors and run the full program in a fixed amount of memory.
But, there is one blocking issue with this approach: XLA compilation is stupidly slow and the framework doesn't provide good options for persisting compiled graphs.
But, based on chats I've had with a JAX dev, there's a good chance these problems are solvable (they just haven't been solved because they haven't mattered much for Google).
In any case, this question could probably be resolved via a few chats with the XLA folks.
What do you think?
from dfdx.
What are your thoughts on using IREE? The IREE runtime supports a diverse range of backends such as Rocm, CUDA, WebGPU, and Metal. Additionally, it would be beneficial if the models could be AOT compiled at Rust compile time, though I'm uncertain if the current architecture of dfdx supports this.
from dfdx.
I've never used IREE but that also seems like a good option.
AOT compilation would be great, too.
Perhaps the critique should be "don't manually support various backends, instead choose an IR and target it instead".
from dfdx.
Related Issues (20)
- Different results when CPU feature is on vs off HOT 2
- Unnecessary loss of precision when computing loss functions HOT 2
- trait TryConcatAlong not satisfied when using constants HOT 1
- Consider helpers for accessing tensors from tuples and input wrappers HOT 1
- Question / clarification regarding heap allocations HOT 2
- Examples or resources for autodiff with 2 networks?
- Bug: `Sequential` macro provide `forward_mut` as `forward`
- Replace explicit features and paths on generated code
- Send/Sync for Device HOT 1
- Add `OUTPUT_PADDING` to `ConvTrans2D`
- Split `TryConcatAlong` into different traits
- Add `Prodigy` optimizer HOT 1
- Run tests with miri HOT 1
- Reduce test sizes HOT 1
- Unclear how to handle error type in `dfdx::nn::LoadFromNpz::load`
- Add `nn::AdaptiveAvgPool2D`
- How does one update one model from another model? HOT 1
- Unable to build with old CUDA version (`CUDA_COMPUTE_CAP = 52`)
- CUDA kernels missing __hmin and __hmax HOT 1
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 dfdx.