ded93a1169
Add the SGD optimizer ( #160 )
...
* Add the nn::optim and some conversion traits.
* Add the backward_step function for SGD.
* Get the SGD optimizer to work and add a test.
* Make the test slighly simpler.
2023-07-13 19:05:44 +01:00
5ee3c95582
Move the variable creation to the variable module. ( #159 )
...
* Move the variable creation to the variable module.
* Make it possible to set a variable.
* Add some basic gradient descent test.
* Get the gradient descent test to work.
2023-07-13 16:55:40 +01:00
6991036bc5
Introduce the variables api used for adjusting parameters during the training loop. ( #158 )
...
* Add the variable api.
* And add a comment.
2023-07-13 14:09:51 +01:00
20599172ac
Add from_iter and arange, use it in the doctests. ( #145 )
2023-07-12 12:03:01 +01:00
fa760759e5
Allow for lazy loading of npz files, use it in llama to reduce memory usage in the cpu version. ( #141 )
2023-07-11 20:22:34 +01:00
64264d97c1
Modular backends ( #138 )
...
* Add some trait to formalize backends.
* Use the generic backend trait.
2023-07-11 11:17:02 +01:00
fba07d6b6b
Merge pull request #127 from LaurentMazare/tensor_indexing
...
`i(..)` indexing sugar (partial).
2023-07-10 19:56:34 +02:00
ef0375d8bc
i(..) indexing sugar (partial).
...
- Only range, and select (no tensor_select)
- No negative indexing
2023-07-10 17:34:04 +02:00
e2807c78a4
Enable the doctests to run with mkl (though they are broken for now). ( #126 )
2023-07-10 16:27:46 +01:00
548b1df7ea
Remove the dependency to blas and use mkl directly. ( #125 )
2023-07-10 15:52:03 +01:00
868743b8b9
Expanding a bit the README
2023-07-10 12:51:37 +02:00
2c3d871b2e
Add a simpler way to specify the dim index for some ops.
2023-07-05 20:22:43 +01:00
a424d95473
Add more of the conv1d op.
2023-07-04 11:15:45 +01:00
cf2789fb81
Move some safetensors bits in the candle-core crate.
2023-07-03 08:37:46 +01:00
c1bbbf94f6
Start refactoring the stride.
2023-06-28 12:57:30 +01:00
8c81a70170
PyTorch like display implementation.
2023-06-27 21:16:35 +01:00
1d504cc6b3
Rework the debug trait.
2023-06-27 19:10:30 +01:00
ca6aa8ff12
Use num-cpus to enable parallelism.
2023-06-27 14:42:26 +01:00
d7f729fb8f
Refactor the hierarchy.
2023-06-27 11:57:27 +02:00
