| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Move some models to candle-transformers so that it's easier to re-use. (#794) | Laurent Mazare | 2023-09-10 | 1 | -485/+0 |
| | | | | | | | | | | * Move some models to candle-transformers so that they can be shared. * Also move falcon. * Move Llama. * Move whisper (partial). | ||||
| * | Add a simple Module trait and implement it for the various nn layers (#500) | Laurent Mazare | 2023-08-18 | 1 | -1/+1 |
| | | | | | | | | * Start adding the module trait. * Use the module trait. * Implement module for qmatmul. | ||||
| * | Use u8 tensors for masks. (#273) | Laurent Mazare | 2023-07-29 | 1 | -1/+1 |
| | | |||||
| * | Softmax numerical stability. (#267) | Laurent Mazare | 2023-07-28 | 1 | -5/+7 |
| | | | | | | * Softmax numerical stability. * Fix the flash-attn test. | ||||
| * | Rename the .r functions to .dims so as to be a bit more explicit. (#220) | Laurent Mazare | 2023-07-22 | 1 | -4/+4 |
| | | |||||
| * | Tensor mutability (#154) | Laurent Mazare | 2023-07-13 | 1 | -3/+3 |
| | | | | | | * Working towards tensor mutability. * Use a ref-cell to provide tensor mutability. | ||||
| * | Use arange in the examples. (#146) | Laurent Mazare | 2023-07-12 | 1 | -2/+1 |
| | | |||||
| * | Remove some dead-code pragmas. (#137) | Laurent Mazare | 2023-07-11 | 1 | -19/+0 |
| | | |||||
| * | VarBuilder path creation (#131) | Laurent Mazare | 2023-07-10 | 1 | -49/+27 |
| | | | | | | | | * Use a struct for the safetensor+routing. * Group the path and the var-builder together. * Fix for the empty path case. | ||||
| * | Move the var-builder in a central place. (#130) | Laurent Mazare | 2023-07-10 | 1 | -60/+2 |
| | | |||||
| * | [nn] Move the Embedding and Activation parts. (#116) | Laurent Mazare | 2023-07-10 | 1 | -29/+5 |
| | | | | | | * Share the Embedding and Activation parts. * Tweak some activations. | ||||
| * | Sketch the candle-nn crate. (#115) | Laurent Mazare | 2023-07-10 | 1 | -79/+34 |
| | | | | | | | | * Sketch the candle-nn crate. * Tweak the cuda dependencies. * More cuda tweaks. | ||||
| * | Use F32 for the reduce ops. (#105) | Laurent Mazare | 2023-07-07 | 1 | -1/+6 |
| | | |||||
| * | Add a KV cache to falcon. (#104) | Laurent Mazare | 2023-07-07 | 1 | -40/+72 |
| | | |||||
| * | Add some caching to the causal mask. (#103) | Laurent Mazare | 2023-07-07 | 1 | -2/+10 |
| | | |||||
| * | Add some text generation pipeline for falcon. (#98) | Laurent Mazare | 2023-07-07 | 1 | -5/+10 |
| | | |||||
| * | Bugfixes. (#97) | Laurent Mazare | 2023-07-06 | 1 | -1/+1 |
| | | |||||
| * | Add the call to dense in the attention layer. (#96) | Laurent Mazare | 2023-07-06 | 1 | -0/+1 |
| | | |||||
| * | Fix some shape issues in falcon. (#95) | Laurent Mazare | 2023-07-06 | 1 | -6/+17 |
| | | | | | | * Fix some shape issues. * Use different dtypes. | ||||
| * | Sketch the Falcon model. (#93) | Laurent Mazare | 2023-07-06 | 1 | -0/+590 |
| * Sketch the Falcon model. * Add more substance to the falcon example. * Falcon (wip). * Falcon (wip again). * Falcon inference. * Get the weights from the api and properly generate the model. * Use the proper model. * Fix the file/revision names. * Fix bias handling. * Recompute the rot embeddings. * Fix the input shape. * Add the release-with-debug profile. * Silly bugfix. * More bugfixes. * Stricter shape checking in matmul. | |||||
 |
index : forks/candle.git | |
| summaryrefslogtreecommitdiff |