Nicolas Patry
|
d7f729fb8f
|
Refactor the hierarchy.
|
2023-06-27 11:57:27 +02:00 |
|
laurent
|
22da2c7e02
|
More f16 and bf16 support.
|
2023-06-26 20:52:01 +01:00 |
|
laurent
|
a31411fd91
|
Start adding f16/bf16 support.
|
2023-06-26 19:37:47 +01:00 |
|
laurent
|
11696e6377
|
Faster model weight loading.
|
2023-06-26 07:40:11 +01:00 |
|
laurent
|
96c098b6cd
|
Remove the unecessary features.
|
2023-06-24 18:15:44 +01:00 |
|
laurent
|
a7f80e258f
|
Read and write npy files.
|
2023-06-24 18:12:10 +01:00 |
|
Nicolas Patry
|
04cf14f35a
|
Moving to `gemm` and adding matmul backprop.
- Tentative `T` operator.
|
2023-06-22 12:37:02 +02:00 |
|
Nicolas Patry
|
9ea220fc6e
|
Fixing tokenizers dep.
|
2023-06-22 12:25:58 +02:00 |
|
Nicolas Patry
|
ce977b489e
|
Adding matmul?
|
2023-06-22 12:25:58 +02:00 |
|
laurent
|
083ced4428
|
Integrate the kernels bits.
|
2023-06-22 09:59:00 +01:00 |
|
laurent
|
7adffafeda
|
Abstract the gradient storage.
|
2023-06-21 14:29:48 +01:00 |
|
laurent
|
9698211d56
|
Add some very basic tensor type.
|
2023-06-19 17:26:50 +01:00 |