From 5f86f59fef77ec060b0020471153611187b14bd3 Mon Sep 17 00:00:00 2001 From: thammuio Date: Mon, 29 Apr 2024 12:27:53 -0400 Subject: [PATCH] fixes --- app/api/routes.py | 2 +- app/chatbot/mistral_7B_instruct/controller.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/app/api/routes.py b/app/api/routes.py index a7b546a..40bae3c 100644 --- a/app/api/routes.py +++ b/app/api/routes.py @@ -23,7 +23,7 @@ def check_api_status() -> dict[str, str]: # Chat # Map the model names to the functions model_methods = { - "llama-2-7B-chat": llama_2_7b_chat + "llama2-7B-chat": llama_2_7b_chat # "llama-2-13B-chat": llama_2_13b_chat, # "Meta-Llama-3-8B-Instruct": llama_3_8b_instruct, # "zephyr-7B-alpha": zephyr_7B_alpha, diff --git a/app/chatbot/mistral_7B_instruct/controller.py b/app/chatbot/mistral_7B_instruct/controller.py index dbd19b7..b01b003 100644 --- a/app/chatbot/mistral_7B_instruct/controller.py +++ b/app/chatbot/mistral_7B_instruct/controller.py @@ -54,7 +54,7 @@ def mistral_7b_instruct(prompt, temperature, max_tokens, selected_vector_db, use context_chunk = get_nearest_chunk_from_pinecone_vectordb(index, vdb_question) # Step 2: Call the relavent Model in Model Serving - MODEL_ACCESS_KEY = get_model_access_key({"name": "llama-2-7B-chat"}) + MODEL_ACCESS_KEY = get_model_access_key({"name": "Mistral-7B-Instruct"}) question = {'prompt': prompt, "temperature": temperature, "max_tokens": max_tokens, "context": context_chunk, "user": user} data = json.dumps({'accessKey': MODEL_ACCESS_KEY, 'request': question})