mirror of
https://github.com/hwchase17/langchain.git
synced 2026-04-23 20:23:59 +00:00
.
This commit is contained in:
@@ -18,7 +18,7 @@ next:
|
||||
- run fine-tune in RetrievalQAChain to do a better side-by-side
|
||||
- there's no HuggingFaceModel?
|
||||
- lora/peft merge
|
||||
- some more error analysis, some more scale in data and maybe model
|
||||
- some more error analysis, some more scale in data and maybe model (esp. for more realistic retrieval setting, larger chunk size, etc.)
|
||||
|
||||
later/not important just noting:
|
||||
- LLMChain doesn't expose raw prompt (and in turn any chain wrapping an LLMChain too), so I have some stateful hackery
|
||||
|
||||
Reference in New Issue
Block a user