github/privateGPT
1
0
Fork 1
mirror of https://github.com/imartinez/privateGPT.git synced 2025-08-15 14:13:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

added request_timeout to ollama, default set to 30.0 in settings.yaml and settings-ollama.yaml

Browse Source
This commit is contained in:
Stephen Gresham 2024-03-15 15:57:13 +11:00
parent 84ad16af80
commit f3aa988e5c
3 changed files with 3 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
private_gpt/components/llm/llm_component.py
View File

@ -131,6 +131,7 @@ class LLMComponent:
temperature=settings.llm.temperature, temperature=settings.llm.temperature,
context_window=settings.llm.context_window, context_window=settings.llm.context_window,
additional_kwargs=settings_kwargs, additional_kwargs=settings_kwargs,
request_timeout=ollama_settings.request_timeout
) )
case "mock": case "mock":
self.llm = MockLLM() self.llm = MockLLM()

1
settings-ollama.yaml
View File

@ -19,6 +19,7 @@ ollama:
top_p: 0.9 # Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. (Default: 0.9) top_p: 0.9 # Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. (Default: 0.9)
repeat_last_n: 64 # Sets how far back for the model to look back to prevent repetition. (Default: 64, 0 = disabled, -1 = num_ctx) repeat_last_n: 64 # Sets how far back for the model to look back to prevent repetition. (Default: 64, 0 = disabled, -1 = num_ctx)
repeat_penalty: 1.2 # Sets how strongly to penalize repetitions. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. (Default: 1.1) repeat_penalty: 1.2 # Sets how strongly to penalize repetitions. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. (Default: 1.1)
request_timeout: 30.0 # Time elapsed until ollama times out the request. Default is 30s. Format is float.
vectorstore: vectorstore:
database: qdrant database: qdrant

1
settings.yaml
View File

@ -86,3 +86,4 @@ ollama:
llm_model: llama2 llm_model: llama2
embedding_model: nomic-embed-text embedding_model: nomic-embed-text
api_base: http://localhost:11434 api_base: http://localhost:11434
request_timeout: 30.0