fixed lama function

.gitignore

@@ -3,7 +3,6 @@ venv/*
 .env
 
 #LLMs
-models/*
 
 #Database
 db/*

main.py

@@ -81,7 +81,7 @@ class VectorDatabaseInitializer:
         text_splitter = CharacterTextSplitter(chunk_size=500, chunk_overlap=50)
         text_chunks = text_splitter.split_documents(raw_documents)
 
-        if selected_embeddings == "openai":
+        if selected_embeddings == "OpenAI":
             embeddings = OpenAIEmbeddings()
         else:
             embeddings = HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L6-v2', model_kwargs={'device': 'cpu'})
@@ -94,7 +94,7 @@ class VectorDatabaseInitializer:
 class LLMHandler:
     @staticmethod
     def initialize_llm(llm_type, temperature, max_tokens):
-        if llm_type == "gpt3.5":
+        if llm_type == "gpt-3.5-turbo":
             llm = ChatOpenAI(model_name="gpt-3.5-turbo", temperature=temperature, max_tokens=max_tokens)
         else:
             llm = CTransformers(model="models/llama-2-7b-chat.ggmlv3.q4_0.bin",
@@ -113,7 +113,7 @@ def handle_user_selection(upload_document, user_query, selected_llms, temperatur
     timestamp = datetime.datetime.now().strftime("%Y%m%d%H%M%S")
 
     vector_db_initializer = VectorDatabaseInitializer()
-    vectordb = vector_db_initializer.initialize_vector_database(raw_documents, selected_embeddings, f'db_{timestamp}')
+    vectordb = vector_db_initializer.initialize_vector_database(raw_documents, selected_embeddings, f'VectorDB/db_{timestamp}')
 
     llm_handler = LLMHandler()
     llm = llm_handler.initialize_llm(selected_llms, temperature, max_tokens)
@@ -169,11 +169,11 @@ with gr.Blocks(theme=gr.Theme.from_hub('finlaymacklon/smooth_slate'), title="LLM
                 query = gr.Textbox(label="Query", placeholder="Type a question and Enter", lines=1)
         with gr.Column(scale=0.1):
             template_input = gr.Textbox(label="Custom Template (optional)", placeholder="Enter your template or leave blank for default", lines=3, scale=1)
-            selected_llm = gr.Radio(["gpt3.5", "Llama"], label="Select LLMs")
+            selected_llm = gr.Radio(choices = ["gpt-3.5-turbo", "llama-2-7b-chat"], label="Select LLMs",value = 'gpt-3.5-turbo')
             temperature = gr.Slider(0.0,1.0, label="temperature", value=0.8)
-            max_tokens = gr.Slider(50, 1500, label="Max_Tokens", value = 50)
-            selected_embeddings = gr.Radio(["openai", "huggingface"], label="Select Embeddings")
-            selected_vector_db = gr.Radio(["Chroma"], label="Select Vector Databases")
+            max_tokens = gr.Slider(50, 1500, label="Max_Tokens", value = 100)
+            selected_embeddings = gr.Radio(choices = ["OpenAI", "MiniLM-L6-v2"], label="Select Embeddings",value = 'OpenAI')
+            selected_vector_db = gr.Radio(choices = ["Chroma"], label="Select Vector Databases",value = 'Chroma')
             clear_vector_button = gr.Button("Clear Vector store")
             clear_vector_output = gr.Textbox(label = "Vector store output")
             clear_history_button = gr.Button("Clear history")

models/llama-2-7b-chat.ggmlv3.q4_0.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8daa9615cce30c259a9555b1cc250d461d1bc69980a274b44d7eda0be78076d8
+size 3791725184