coreylowman/dfdx
Deep learning in Rust, with shape checked tensors and neural networks
RustNOASSERTION
Issues
- 0
Basic operators moving tensor variables
#929 opened by otavio-silva - 0
Safetensor panics instead of returning Err
#928 opened by opfromthestart - 1
Kernels written in rust-gpu
#925 opened by LegNeato - 2
Non-trainable parameters?
#921 opened by hovinen - 0
Tests failing on HEAD /w cargo 1.78.0
#926 opened by twitchyliquid64 - 0
can not perform tensor operations (add/sub/mul/...) on tensors with non float datatype
#924 opened by lucomotive - 1
- 1
Run tests with miri
#897 opened by swfsql - 1
Reduce test sizes
#898 opened by swfsql - 1
Add `Prodigy` optimizer
#894 opened by swfsql - 1
How does one update one model from another model?
#905 opened by hovinen - 3
OpenXLA Support
#908 opened by vyeevani - 7
Please add a split method, to iterate over subtensors of the given tensor
#825 opened by emchristiansen - 1
CUDA kernels missing __hmin and __hmax
#910 opened by opfromthestart - 3
`TapeGlobal` and thoughts on the tape variants
#843 opened by emchristiansen - 0
- 0
Add `nn::AdaptiveAvgPool2D`
#903 opened by quietlychris - 0
- 0
Split `TryConcatAlong` into different traits
#891 opened by swfsql - 0
Replace explicit features and paths on generated code
#885 opened by swfsql - 0
Add `OUTPUT_PADDING` to `ConvTrans2D`
#889 opened by swfsql - 1
Send/Sync for Device
#888 opened by blogle - 2
- 0
- 0
Examples or resources for autodiff with 2 networks?
#880 opened by Cobrand - 1
Deriving Shape for C like enums
#856 opened by ZacharyKamerling - 1
Is there an equivalent to torch.masked_select?
#865 opened by jackerschott - 2
- 1
- 2
Different results when CPU feature is on vs off
#871 opened by jafioti - 0
Alloc zero size memory on old model GPU may fail.
#870 opened by optman - 1
Impossible to create a module with a parameter that lacks a const shape?
#839 opened by emchristiansen - 1
- 2
Device doesn't cover all traits from Cpu/Cuda
#866 opened by jackerschott - 12
ConcatAlong is very slow
#867 opened by opfromthestart - 1
Expose AddDim to public API
#859 opened by LeoDog896 - 1
Convenience of mixed tensor scalar operations
#862 opened by jackerschott - 1
Question: How to load checkpoints?
#861 opened by rjzak - 1
Grouped Linear layers
#857 opened by opfromthestart - 3
How to calculate loss to maximize a value
#850 opened by opfromthestart - 6
Differing behavior between Pytorch and dfdx
#848 opened by opfromthestart - 1
impl `Merge` for `NoneTape` for `OwnedTape` and `Arc<Mutex<OwnedTape>>`
#836 opened by emchristiansen - 2
Consider adding a `Tape` type where the tape is optional, e.g. `Option<Arc<Mutex<OwnedTape<_, _>>>>`
#837 opened by emchristiansen - 0
Bug: `as_vec` crashes for empty Tensors
#844 opened by emchristiansen - 2
The merge semantics for `Arc<Mutex<OwnedTape<_, _>>>` seem a bit unintuitive
#841 opened by emchristiansen - 1
- 4
`ArcTape`: Add another `Tape` impl?
#833 opened by emchristiansen - 7
- 1
Add Gradient scaler
#832 opened by coreylowman - 2