12 Commits (a3177ab1404820e17bcfe3ecb1ae6de917c74672)

Author SHA1 Message Date
  Martin Evans ea523d2e2a Renamed `llama_sample_temperature` to `llama_sample_temp`, Mirroring the same change made in llama.cpp 1 year ago
  Martin Evans 835958398c - Removed the object wrappers and configurable pipeline, they can be better written in code. 1 year ago
  Martin Evans 33358124db Initial pass at a new sampling pipeline 1 year ago
  Martin Evans d743516070 - Added support for the MinP sampler 2 years ago
  Martin Evans e81b3023d5 Rewritten sampling API to be accessed through the `LLamaTokenDataArray` object 2 years ago
  Martin Evans 3c5547b2b7 Reduced some uses of `NativeApi` in `BatchedDecoding` by adding some helper methods 2 years ago
  Martin Evans cf4754db44 Removed unnecessary parameters from some low level sampler methods 2 years ago
  zombieguy 45b01d5a78 Improved type conversion 2 years ago
  zombieguy 10f88ebd0e
Potential fix for .Net Framework issues (#103) 2 years ago
  Martin Evans ec49bdd6eb - Most importantly: Fixed issue in `SamplingApi`, `Memory` was pinned, but never unpinned! 2 years ago
  Yaohui Liu 264fb9a706
refactor: LLamaModel and LLamaExecutor. 2 years ago
  Yaohui Liu 5a79edeb51
feat: add the framework and basic usages. 2 years ago